init

.gitignore

@@ -0,0 +1,18 @@
+## Ignore some files and folders for copyright and other reasons. 
+
+*.model
+*.model.base
+[Mm]odel_speech/
+
+__pycache__
+*.wav
+*.model_yaml
+Test_Report_*
+data/*
+dataset
+data_pinyin.txt
+*.trn
+testClient.py
+
+webapi/
+save_models/

Dockerfile

@@ -0,0 +1,45 @@
+# Ubuntu 20.04
+FROM ubuntu:focal
+
+# 切换默认shell为bash
+SHELL ["/bin/bash", "-c"]
+
+ADD ./ /asrt_server
+
+WORKDIR /asrt_server
+
+# 最小化源，缩短apt update时间(ca-certificates必须先安装才支持换tsinghua源)
+RUN echo 'deb http://archive.ubuntu.com/ubuntu/ focal main restricted' > /etc/apt/sources.list
+
+RUN apt update && apt install -y ca-certificates
+
+RUN echo $'\
+deb http://mirrors.aliyun.com/ubuntu/ focal main restricted universe multiverse\
+\n# deb-src http://mirrors.aliyun.com/ubuntu/ focal main restricted universe multiverse\n\
+deb http://mirrors.aliyun.com/ubuntu/ focal-security main restricted universe multiverse\
+\n# deb-src http://mirrors.aliyun.com/ubuntu/ focal-security main restricted universe multiverse\n\
+deb http://mirrors.aliyun.com/ubuntu/ focal-updates main restricted universe multiverse\
+\n# deb-src http://mirrors.aliyun.com/ubuntu/ focal-updates main restricted universe multiverse\n\
+deb http://mirrors.aliyun.com/ubuntu/ focal-proposed main restricted universe multiverse\
+\n# deb-src http://mirrors.aliyun.com/ubuntu/ focal-proposed main restricted universe multiverse\n\
+deb http://mirrors.aliyun.com/ubuntu/ focal-backports main restricted universe multiverse\
+\n# deb-src http://mirrors.aliyun.com/ubuntu/ focal-backports main restricted universe multiverse'\
+> /etc/apt/sources.list
+
+RUN apt update && apt install -y python3 python3-pip 
+
+RUN pip3 config set global.index-url https://pypi.tuna.tsinghua.edu.cn/simple
+
+RUN pip3 install wave scipy matplotlib tensorflow-cpu==2.5.3 numpy==1.19.2 requests flask waitress grpcio==1.34.0 grpcio-tools==1.34.0 protobuf
+
+RUN echo $'cd /asrt_server \n python3 asrserver_http.py & \n python3 asrserver_grpc.py' > /asrt_server/start.sh && chmod +x /asrt_server/start.sh
+
+# refer: https://docs.docker.com/engine/reference/builder/#expose
+EXPOSE 20001/tcp 20002/tcp
+
+ENTRYPOINT ["/bin/bash", "/asrt_server/start.sh"]
+
+# https://docs.docker.com/engine/reference/commandline/build/#options
+# docker build --progress plain --rm --build-arg TAG=1.3.0 --tag asrt/api_server:1.3.0 .
+# https://docs.docker.com/engine/reference/commandline/run/#options
+# docker run --rm -it -p 20001:20001 -p 20002:20002 --name asrt -d asrt/api_server:1.3.0

LICENSE

@@ -0,0 +1,674 @@
+                    GNU GENERAL PUBLIC LICENSE
+                       Version 3, 29 June 2007
+
+ Copyright (C) 2007 Free Software Foundation, Inc. <http://fsf.org/>
+ Everyone is permitted to copy and distribute verbatim copies
+ of this license document, but changing it is not allowed.
+
+                            Preamble
+
+  The GNU General Public License is a free, copyleft license for
+software and other kinds of works.
+
+  The licenses for most software and other practical works are designed
+to take away your freedom to share and change the works.  By contrast,
+the GNU General Public License is intended to guarantee your freedom to
+share and change all versions of a program--to make sure it remains free
+software for all its users.  We, the Free Software Foundation, use the
+GNU General Public License for most of our software; it applies also to
+any other work released this way by its authors.  You can apply it to
+your programs, too.
+
+  When we speak of free software, we are referring to freedom, not
+price.  Our General Public Licenses are designed to make sure that you
+have the freedom to distribute copies of free software (and charge for
+them if you wish), that you receive source code or can get it if you
+want it, that you can change the software or use pieces of it in new
+free programs, and that you know you can do these things.
+
+  To protect your rights, we need to prevent others from denying you
+these rights or asking you to surrender the rights.  Therefore, you have
+certain responsibilities if you distribute copies of the software, or if
+you modify it: responsibilities to respect the freedom of others.
+
+  For example, if you distribute copies of such a program, whether
+gratis or for a fee, you must pass on to the recipients the same
+freedoms that you received.  You must make sure that they, too, receive
+or can get the source code.  And you must show them these terms so they
+know their rights.
+
+  Developers that use the GNU GPL protect your rights with two steps:
+(1) assert copyright on the software, and (2) offer you this License
+giving you legal permission to copy, distribute and/or modify it.
+
+  For the developers' and authors' protection, the GPL clearly explains
+that there is no warranty for this free software.  For both users' and
+authors' sake, the GPL requires that modified versions be marked as
+changed, so that their problems will not be attributed erroneously to
+authors of previous versions.
+
+  Some devices are designed to deny users access to install or run
+modified versions of the software inside them, although the manufacturer
+can do so.  This is fundamentally incompatible with the aim of
+protecting users' freedom to change the software.  The systematic
+pattern of such abuse occurs in the area of products for individuals to
+use, which is precisely where it is most unacceptable.  Therefore, we
+have designed this version of the GPL to prohibit the practice for those
+products.  If such problems arise substantially in other domains, we
+stand ready to extend this provision to those domains in future versions
+of the GPL, as needed to protect the freedom of users.
+
+  Finally, every program is threatened constantly by software patents.
+States should not allow patents to restrict development and use of
+software on general-purpose computers, but in those that do, we wish to
+avoid the special danger that patents applied to a free program could
+make it effectively proprietary.  To prevent this, the GPL assures that
+patents cannot be used to render the program non-free.
+
+  The precise terms and conditions for copying, distribution and
+modification follow.
+
+                       TERMS AND CONDITIONS
+
+  0. Definitions.
+
+  "This License" refers to version 3 of the GNU General Public License.
+
+  "Copyright" also means copyright-like laws that apply to other kinds of
+works, such as semiconductor masks.
+
+  "The Program" refers to any copyrightable work licensed under this
+License.  Each licensee is addressed as "you".  "Licensees" and
+"recipients" may be individuals or organizations.
+
+  To "modify" a work means to copy from or adapt all or part of the work
+in a fashion requiring copyright permission, other than the making of an
+exact copy.  The resulting work is called a "modified version" of the
+earlier work or a work "based on" the earlier work.
+
+  A "covered work" means either the unmodified Program or a work based
+on the Program.
+
+  To "propagate" a work means to do anything with it that, without
+permission, would make you directly or secondarily liable for
+infringement under applicable copyright law, except executing it on a
+computer or modifying a private copy.  Propagation includes copying,
+distribution (with or without modification), making available to the
+public, and in some countries other activities as well.
+
+  To "convey" a work means any kind of propagation that enables other
+parties to make or receive copies.  Mere interaction with a user through
+a computer network, with no transfer of a copy, is not conveying.
+
+  An interactive user interface displays "Appropriate Legal Notices"
+to the extent that it includes a convenient and prominently visible
+feature that (1) displays an appropriate copyright notice, and (2)
+tells the user that there is no warranty for the work (except to the
+extent that warranties are provided), that licensees may convey the
+work under this License, and how to view a copy of this License.  If
+the interface presents a list of user commands or options, such as a
+menu, a prominent item in the list meets this criterion.
+
+  1. Source Code.
+
+  The "source code" for a work means the preferred form of the work
+for making modifications to it.  "Object code" means any non-source
+form of a work.
+
+  A "Standard Interface" means an interface that either is an official
+standard defined by a recognized standards body, or, in the case of
+interfaces specified for a particular programming language, one that
+is widely used among developers working in that language.
+
+  The "System Libraries" of an executable work include anything, other
+than the work as a whole, that (a) is included in the normal form of
+packaging a Major Component, but which is not part of that Major
+Component, and (b) serves only to enable use of the work with that
+Major Component, or to implement a Standard Interface for which an
+implementation is available to the public in source code form.  A
+"Major Component", in this context, means a major essential component
+(kernel, window system, and so on) of the specific operating system
+(if any) on which the executable work runs, or a compiler used to
+produce the work, or an object code interpreter used to run it.
+
+  The "Corresponding Source" for a work in object code form means all
+the source code needed to generate, install, and (for an executable
+work) run the object code and to modify the work, including scripts to
+control those activities.  However, it does not include the work's
+System Libraries, or general-purpose tools or generally available free
+programs which are used unmodified in performing those activities but
+which are not part of the work.  For example, Corresponding Source
+includes interface definition files associated with source files for
+the work, and the source code for shared libraries and dynamically
+linked subprograms that the work is specifically designed to require,
+such as by intimate data communication or control flow between those
+subprograms and other parts of the work.
+
+  The Corresponding Source need not include anything that users
+can regenerate automatically from other parts of the Corresponding
+Source.
+
+  The Corresponding Source for a work in source code form is that
+same work.
+
+  2. Basic Permissions.
+
+  All rights granted under this License are granted for the term of
+copyright on the Program, and are irrevocable provided the stated
+conditions are met.  This License explicitly affirms your unlimited
+permission to run the unmodified Program.  The output from running a
+covered work is covered by this License only if the output, given its
+content, constitutes a covered work.  This License acknowledges your
+rights of fair use or other equivalent, as provided by copyright law.
+
+  You may make, run and propagate covered works that you do not
+convey, without conditions so long as your license otherwise remains
+in force.  You may convey covered works to others for the sole purpose
+of having them make modifications exclusively for you, or provide you
+with facilities for running those works, provided that you comply with
+the terms of this License in conveying all material for which you do
+not control copyright.  Those thus making or running the covered works
+for you must do so exclusively on your behalf, under your direction
+and control, on terms that prohibit them from making any copies of
+your copyrighted material outside their relationship with you.
+
+  Conveying under any other circumstances is permitted solely under
+the conditions stated below.  Sublicensing is not allowed; section 10
+makes it unnecessary.
+
+  3. Protecting Users' Legal Rights From Anti-Circumvention Law.
+
+  No covered work shall be deemed part of an effective technological
+measure under any applicable law fulfilling obligations under article
+11 of the WIPO copyright treaty adopted on 20 December 1996, or
+similar laws prohibiting or restricting circumvention of such
+measures.
+
+  When you convey a covered work, you waive any legal power to forbid
+circumvention of technological measures to the extent such circumvention
+is effected by exercising rights under this License with respect to
+the covered work, and you disclaim any intention to limit operation or
+modification of the work as a means of enforcing, against the work's
+users, your or third parties' legal rights to forbid circumvention of
+technological measures.
+
+  4. Conveying Verbatim Copies.
+
+  You may convey verbatim copies of the Program's source code as you
+receive it, in any medium, provided that you conspicuously and
+appropriately publish on each copy an appropriate copyright notice;
+keep intact all notices stating that this License and any
+non-permissive terms added in accord with section 7 apply to the code;
+keep intact all notices of the absence of any warranty; and give all
+recipients a copy of this License along with the Program.
+
+  You may charge any price or no price for each copy that you convey,
+and you may offer support or warranty protection for a fee.
+
+  5. Conveying Modified Source Versions.
+
+  You may convey a work based on the Program, or the modifications to
+produce it from the Program, in the form of source code under the
+terms of section 4, provided that you also meet all of these conditions:
+
+    a) The work must carry prominent notices stating that you modified
+    it, and giving a relevant date.
+
+    b) The work must carry prominent notices stating that it is
+    released under this License and any conditions added under section
+    7.  This requirement modifies the requirement in section 4 to
+    "keep intact all notices".
+
+    c) You must license the entire work, as a whole, under this
+    License to anyone who comes into possession of a copy.  This
+    License will therefore apply, along with any applicable section 7
+    additional terms, to the whole of the work, and all its parts,
+    regardless of how they are packaged.  This License gives no
+    permission to license the work in any other way, but it does not
+    invalidate such permission if you have separately received it.
+
+    d) If the work has interactive user interfaces, each must display
+    Appropriate Legal Notices; however, if the Program has interactive
+    interfaces that do not display Appropriate Legal Notices, your
+    work need not make them do so.
+
+  A compilation of a covered work with other separate and independent
+works, which are not by their nature extensions of the covered work,
+and which are not combined with it such as to form a larger program,
+in or on a volume of a storage or distribution medium, is called an
+"aggregate" if the compilation and its resulting copyright are not
+used to limit the access or legal rights of the compilation's users
+beyond what the individual works permit.  Inclusion of a covered work
+in an aggregate does not cause this License to apply to the other
+parts of the aggregate.
+
+  6. Conveying Non-Source Forms.
+
+  You may convey a covered work in object code form under the terms
+of sections 4 and 5, provided that you also convey the
+machine-readable Corresponding Source under the terms of this License,
+in one of these ways:
+
+    a) Convey the object code in, or embodied in, a physical product
+    (including a physical distribution medium), accompanied by the
+    Corresponding Source fixed on a durable physical medium
+    customarily used for software interchange.
+
+    b) Convey the object code in, or embodied in, a physical product
+    (including a physical distribution medium), accompanied by a
+    written offer, valid for at least three years and valid for as
+    long as you offer spare parts or customer support for that product
+    model, to give anyone who possesses the object code either (1) a
+    copy of the Corresponding Source for all the software in the
+    product that is covered by this License, on a durable physical
+    medium customarily used for software interchange, for a price no
+    more than your reasonable cost of physically performing this
+    conveying of source, or (2) access to copy the
+    Corresponding Source from a network server at no charge.
+
+    c) Convey individual copies of the object code with a copy of the
+    written offer to provide the Corresponding Source.  This
+    alternative is allowed only occasionally and noncommercially, and
+    only if you received the object code with such an offer, in accord
+    with subsection 6b.
+
+    d) Convey the object code by offering access from a designated
+    place (gratis or for a charge), and offer equivalent access to the
+    Corresponding Source in the same way through the same place at no
+    further charge.  You need not require recipients to copy the
+    Corresponding Source along with the object code.  If the place to
+    copy the object code is a network server, the Corresponding Source
+    may be on a different server (operated by you or a third party)
+    that supports equivalent copying facilities, provided you maintain
+    clear directions next to the object code saying where to find the
+    Corresponding Source.  Regardless of what server hosts the
+    Corresponding Source, you remain obligated to ensure that it is
+    available for as long as needed to satisfy these requirements.
+
+    e) Convey the object code using peer-to-peer transmission, provided
+    you inform other peers where the object code and Corresponding
+    Source of the work are being offered to the general public at no
+    charge under subsection 6d.
+
+  A separable portion of the object code, whose source code is excluded
+from the Corresponding Source as a System Library, need not be
+included in conveying the object code work.
+
+  A "User Product" is either (1) a "consumer product", which means any
+tangible personal property which is normally used for personal, family,
+or household purposes, or (2) anything designed or sold for incorporation
+into a dwelling.  In determining whether a product is a consumer product,
+doubtful cases shall be resolved in favor of coverage.  For a particular
+product received by a particular user, "normally used" refers to a
+typical or common use of that class of product, regardless of the status
+of the particular user or of the way in which the particular user
+actually uses, or expects or is expected to use, the product.  A product
+is a consumer product regardless of whether the product has substantial
+commercial, industrial or non-consumer uses, unless such uses represent
+the only significant mode of use of the product.
+
+  "Installation Information" for a User Product means any methods,
+procedures, authorization keys, or other information required to install
+and execute modified versions of a covered work in that User Product from
+a modified version of its Corresponding Source.  The information must
+suffice to ensure that the continued functioning of the modified object
+code is in no case prevented or interfered with solely because
+modification has been made.
+
+  If you convey an object code work under this section in, or with, or
+specifically for use in, a User Product, and the conveying occurs as
+part of a transaction in which the right of possession and use of the
+User Product is transferred to the recipient in perpetuity or for a
+fixed term (regardless of how the transaction is characterized), the
+Corresponding Source conveyed under this section must be accompanied
+by the Installation Information.  But this requirement does not apply
+if neither you nor any third party retains the ability to install
+modified object code on the User Product (for example, the work has
+been installed in ROM).
+
+  The requirement to provide Installation Information does not include a
+requirement to continue to provide support service, warranty, or updates
+for a work that has been modified or installed by the recipient, or for
+the User Product in which it has been modified or installed.  Access to a
+network may be denied when the modification itself materially and
+adversely affects the operation of the network or violates the rules and
+protocols for communication across the network.
+
+  Corresponding Source conveyed, and Installation Information provided,
+in accord with this section must be in a format that is publicly
+documented (and with an implementation available to the public in
+source code form), and must require no special password or key for
+unpacking, reading or copying.
+
+  7. Additional Terms.
+
+  "Additional permissions" are terms that supplement the terms of this
+License by making exceptions from one or more of its conditions.
+Additional permissions that are applicable to the entire Program shall
+be treated as though they were included in this License, to the extent
+that they are valid under applicable law.  If additional permissions
+apply only to part of the Program, that part may be used separately
+under those permissions, but the entire Program remains governed by
+this License without regard to the additional permissions.
+
+  When you convey a copy of a covered work, you may at your option
+remove any additional permissions from that copy, or from any part of
+it.  (Additional permissions may be written to require their own
+removal in certain cases when you modify the work.)  You may place
+additional permissions on material, added by you to a covered work,
+for which you have or can give appropriate copyright permission.
+
+  Notwithstanding any other provision of this License, for material you
+add to a covered work, you may (if authorized by the copyright holders of
+that material) supplement the terms of this License with terms:
+
+    a) Disclaiming warranty or limiting liability differently from the
+    terms of sections 15 and 16 of this License; or
+
+    b) Requiring preservation of specified reasonable legal notices or
+    author attributions in that material or in the Appropriate Legal
+    Notices displayed by works containing it; or
+
+    c) Prohibiting misrepresentation of the origin of that material, or
+    requiring that modified versions of such material be marked in
+    reasonable ways as different from the original version; or
+
+    d) Limiting the use for publicity purposes of names of licensors or
+    authors of the material; or
+
+    e) Declining to grant rights under trademark law for use of some
+    trade names, trademarks, or service marks; or
+
+    f) Requiring indemnification of licensors and authors of that
+    material by anyone who conveys the material (or modified versions of
+    it) with contractual assumptions of liability to the recipient, for
+    any liability that these contractual assumptions directly impose on
+    those licensors and authors.
+
+  All other non-permissive additional terms are considered "further
+restrictions" within the meaning of section 10.  If the Program as you
+received it, or any part of it, contains a notice stating that it is
+governed by this License along with a term that is a further
+restriction, you may remove that term.  If a license document contains
+a further restriction but permits relicensing or conveying under this
+License, you may add to a covered work material governed by the terms
+of that license document, provided that the further restriction does
+not survive such relicensing or conveying.
+
+  If you add terms to a covered work in accord with this section, you
+must place, in the relevant source files, a statement of the
+additional terms that apply to those files, or a notice indicating
+where to find the applicable terms.
+
+  Additional terms, permissive or non-permissive, may be stated in the
+form of a separately written license, or stated as exceptions;
+the above requirements apply either way.
+
+  8. Termination.
+
+  You may not propagate or modify a covered work except as expressly
+provided under this License.  Any attempt otherwise to propagate or
+modify it is void, and will automatically terminate your rights under
+this License (including any patent licenses granted under the third
+paragraph of section 11).
+
+  However, if you cease all violation of this License, then your
+license from a particular copyright holder is reinstated (a)
+provisionally, unless and until the copyright holder explicitly and
+finally terminates your license, and (b) permanently, if the copyright
+holder fails to notify you of the violation by some reasonable means
+prior to 60 days after the cessation.
+
+  Moreover, your license from a particular copyright holder is
+reinstated permanently if the copyright holder notifies you of the
+violation by some reasonable means, this is the first time you have
+received notice of violation of this License (for any work) from that
+copyright holder, and you cure the violation prior to 30 days after
+your receipt of the notice.
+
+  Termination of your rights under this section does not terminate the
+licenses of parties who have received copies or rights from you under
+this License.  If your rights have been terminated and not permanently
+reinstated, you do not qualify to receive new licenses for the same
+material under section 10.
+
+  9. Acceptance Not Required for Having Copies.
+
+  You are not required to accept this License in order to receive or
+run a copy of the Program.  Ancillary propagation of a covered work
+occurring solely as a consequence of using peer-to-peer transmission
+to receive a copy likewise does not require acceptance.  However,
+nothing other than this License grants you permission to propagate or
+modify any covered work.  These actions infringe copyright if you do
+not accept this License.  Therefore, by modifying or propagating a
+covered work, you indicate your acceptance of this License to do so.
+
+  10. Automatic Licensing of Downstream Recipients.
+
+  Each time you convey a covered work, the recipient automatically
+receives a license from the original licensors, to run, modify and
+propagate that work, subject to this License.  You are not responsible
+for enforcing compliance by third parties with this License.
+
+  An "entity transaction" is a transaction transferring control of an
+organization, or substantially all assets of one, or subdividing an
+organization, or merging organizations.  If propagation of a covered
+work results from an entity transaction, each party to that
+transaction who receives a copy of the work also receives whatever
+licenses to the work the party's predecessor in interest had or could
+give under the previous paragraph, plus a right to possession of the
+Corresponding Source of the work from the predecessor in interest, if
+the predecessor has it or can get it with reasonable efforts.
+
+  You may not impose any further restrictions on the exercise of the
+rights granted or affirmed under this License.  For example, you may
+not impose a license fee, royalty, or other charge for exercise of
+rights granted under this License, and you may not initiate litigation
+(including a cross-claim or counterclaim in a lawsuit) alleging that
+any patent claim is infringed by making, using, selling, offering for
+sale, or importing the Program or any portion of it.
+
+  11. Patents.
+
+  A "contributor" is a copyright holder who authorizes use under this
+License of the Program or a work on which the Program is based.  The
+work thus licensed is called the contributor's "contributor version".
+
+  A contributor's "essential patent claims" are all patent claims
+owned or controlled by the contributor, whether already acquired or
+hereafter acquired, that would be infringed by some manner, permitted
+by this License, of making, using, or selling its contributor version,
+but do not include claims that would be infringed only as a
+consequence of further modification of the contributor version.  For
+purposes of this definition, "control" includes the right to grant
+patent sublicenses in a manner consistent with the requirements of
+this License.
+
+  Each contributor grants you a non-exclusive, worldwide, royalty-free
+patent license under the contributor's essential patent claims, to
+make, use, sell, offer for sale, import and otherwise run, modify and
+propagate the contents of its contributor version.
+
+  In the following three paragraphs, a "patent license" is any express
+agreement or commitment, however denominated, not to enforce a patent
+(such as an express permission to practice a patent or covenant not to
+sue for patent infringement).  To "grant" such a patent license to a
+party means to make such an agreement or commitment not to enforce a
+patent against the party.
+
+  If you convey a covered work, knowingly relying on a patent license,
+and the Corresponding Source of the work is not available for anyone
+to copy, free of charge and under the terms of this License, through a
+publicly available network server or other readily accessible means,
+then you must either (1) cause the Corresponding Source to be so
+available, or (2) arrange to deprive yourself of the benefit of the
+patent license for this particular work, or (3) arrange, in a manner
+consistent with the requirements of this License, to extend the patent
+license to downstream recipients.  "Knowingly relying" means you have
+actual knowledge that, but for the patent license, your conveying the
+covered work in a country, or your recipient's use of the covered work
+in a country, would infringe one or more identifiable patents in that
+country that you have reason to believe are valid.
+
+  If, pursuant to or in connection with a single transaction or
+arrangement, you convey, or propagate by procuring conveyance of, a
+covered work, and grant a patent license to some of the parties
+receiving the covered work authorizing them to use, propagate, modify
+or convey a specific copy of the covered work, then the patent license
+you grant is automatically extended to all recipients of the covered
+work and works based on it.
+
+  A patent license is "discriminatory" if it does not include within
+the scope of its coverage, prohibits the exercise of, or is
+conditioned on the non-exercise of one or more of the rights that are
+specifically granted under this License.  You may not convey a covered
+work if you are a party to an arrangement with a third party that is
+in the business of distributing software, under which you make payment
+to the third party based on the extent of your activity of conveying
+the work, and under which the third party grants, to any of the
+parties who would receive the covered work from you, a discriminatory
+patent license (a) in connection with copies of the covered work
+conveyed by you (or copies made from those copies), or (b) primarily
+for and in connection with specific products or compilations that
+contain the covered work, unless you entered into that arrangement,
+or that patent license was granted, prior to 28 March 2007.
+
+  Nothing in this License shall be construed as excluding or limiting
+any implied license or other defenses to infringement that may
+otherwise be available to you under applicable patent law.
+
+  12. No Surrender of Others' Freedom.
+
+  If conditions are imposed on you (whether by court order, agreement or
+otherwise) that contradict the conditions of this License, they do not
+excuse you from the conditions of this License.  If you cannot convey a
+covered work so as to satisfy simultaneously your obligations under this
+License and any other pertinent obligations, then as a consequence you may
+not convey it at all.  For example, if you agree to terms that obligate you
+to collect a royalty for further conveying from those to whom you convey
+the Program, the only way you could satisfy both those terms and this
+License would be to refrain entirely from conveying the Program.
+
+  13. Use with the GNU Affero General Public License.
+
+  Notwithstanding any other provision of this License, you have
+permission to link or combine any covered work with a work licensed
+under version 3 of the GNU Affero General Public License into a single
+combined work, and to convey the resulting work.  The terms of this
+License will continue to apply to the part which is the covered work,
+but the special requirements of the GNU Affero General Public License,
+section 13, concerning interaction through a network will apply to the
+combination as such.
+
+  14. Revised Versions of this License.
+
+  The Free Software Foundation may publish revised and/or new versions of
+the GNU General Public License from time to time.  Such new versions will
+be similar in spirit to the present version, but may differ in detail to
+address new problems or concerns.
+
+  Each version is given a distinguishing version number.  If the
+Program specifies that a certain numbered version of the GNU General
+Public License "or any later version" applies to it, you have the
+option of following the terms and conditions either of that numbered
+version or of any later version published by the Free Software
+Foundation.  If the Program does not specify a version number of the
+GNU General Public License, you may choose any version ever published
+by the Free Software Foundation.
+
+  If the Program specifies that a proxy can decide which future
+versions of the GNU General Public License can be used, that proxy's
+public statement of acceptance of a version permanently authorizes you
+to choose that version for the Program.
+
+  Later license versions may give you additional or different
+permissions.  However, no additional obligations are imposed on any
+author or copyright holder as a result of your choosing to follow a
+later version.
+
+  15. Disclaimer of Warranty.
+
+  THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY
+APPLICABLE LAW.  EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT
+HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY
+OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO,
+THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
+PURPOSE.  THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM
+IS WITH YOU.  SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF
+ALL NECESSARY SERVICING, REPAIR OR CORRECTION.
+
+  16. Limitation of Liability.
+
+  IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING
+WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MODIFIES AND/OR CONVEYS
+THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY
+GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE
+USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF
+DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD
+PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS),
+EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF
+SUCH DAMAGES.
+
+  17. Interpretation of Sections 15 and 16.
+
+  If the disclaimer of warranty and limitation of liability provided
+above cannot be given local legal effect according to their terms,
+reviewing courts shall apply local law that most closely approximates
+an absolute waiver of all civil liability in connection with the
+Program, unless a warranty or assumption of liability accompanies a
+copy of the Program in return for a fee.
+
+                     END OF TERMS AND CONDITIONS
+
+            How to Apply These Terms to Your New Programs
+
+  If you develop a new program, and you want it to be of the greatest
+possible use to the public, the best way to achieve this is to make it
+free software which everyone can redistribute and change under these terms.
+
+  To do so, attach the following notices to the program.  It is safest
+to attach them to the start of each source file to most effectively
+state the exclusion of warranty; and each file should have at least
+the "copyright" line and a pointer to where the full notice is found.
+
+    {one line to give the program's name and a brief idea of what it does.}
+    Copyright (C) {year}  {name of author}
+
+    This program is free software: you can redistribute it and/or modify
+    it under the terms of the GNU General Public License as published by
+    the Free Software Foundation, either version 3 of the License, or
+    (at your option) any later version.
+
+    This program is distributed in the hope that it will be useful,
+    but WITHOUT ANY WARRANTY; without even the implied warranty of
+    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+    GNU General Public License for more details.
+
+    You should have received a copy of the GNU General Public License
+    along with this program.  If not, see <http://www.gnu.org/licenses/>.
+
+Also add information on how to contact you by electronic and paper mail.
+
+  If the program does terminal interaction, make it output a short
+notice like this when it starts in an interactive mode:
+
+    ASRT_SpeechRecognition  Copyright (C) 2017  nl8590687
+    This program comes with ABSOLUTELY NO WARRANTY; for details type `show w'.
+    This is free software, and you are welcome to redistribute it
+    under certain conditions; type `show c' for details.
+
+The hypothetical commands `show w' and `show c' should show the appropriate
+parts of the General Public License.  Of course, your program's commands
+might be different; for a GUI interface, you would use an "about box".
+
+  You should also get your employer (if you work as a programmer) or school,
+if any, to sign a "copyright disclaimer" for the program, if necessary.
+For more information on this, and how to apply and follow the GNU GPL, see
+<http://www.gnu.org/licenses/>.
+
+  The GNU General Public License does not permit incorporating your program
+into proprietary programs.  If your program is a subroutine library, you
+may consider it more useful to permit linking proprietary applications with
+the library.  If this is what you want to do, use the GNU Lesser General
+Public License instead of this License.  But first, please read
+<http://www.gnu.org/philosophy/why-not-lgpl.html>.

README_CH.md

@@ -0,0 +1,237 @@
+![](assets/asrt_title_header.png)
+
+[![GPL-3.0 Licensed](https://img.shields.io/badge/License-GPL3.0-blue.svg?style=flat)](https://opensource.org/licenses/GPL-3.0) 
+[![Stars](https://img.shields.io/github/stars/nl8590687/ASRT_SpeechRecognition)](https://github.com/nl8590687/ASRT_SpeechRecognition) 
+[![TensorFlow Version](https://img.shields.io/badge/Tensorflow-2.5+-blue.svg)](https://www.tensorflow.org/) 
+[![Python Version](https://img.shields.io/badge/Python-3.7+-blue.svg)](https://www.python.org/) 
+[![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.5808434.svg)](https://doi.org/10.5281/zenodo.5808434)
+
+ASRT是一个基于深度学习的中文语音识别系统，如果您觉得喜欢，请点一个 **"Star"** 吧~
+
+**ReadMe Language** | 中文版 | [English](https://github.com/nl8590687/ASRT_SpeechRecognition/blob/master/README_EN.md) |
+
+[**ASRT项目主页**](https://asrt.ailemon.net/) | 
+[**发布版下载**](https://wiki.ailemon.net/docs/asrt-doc/download) | 
+[**查看本项目的Wiki文档**](https://wiki.ailemon.net/docs/asrt-doc) | 
+[**实用效果体验Demo**](https://asrt.ailemon.net/demo) | 
+[**打赏作者**](https://wiki.ailemon.net/docs/asrt-doc/asrt-doc-1deo9u61unti9)
+
+如果程序运行期间或使用中有什么问题，可以及时在issue中提出来，我将尽快做出答复。本项目作者交流QQ群：**894112051** ，加微信群请先加AI柠檬微信号：**ailemon-me** ，并备注“ASRT语音识别”
+
+<center><img src="https://res.ailemon.net/common/ailemon-me-wechat-qrcode.jpg?x-oss-process=style/ailemon-blog-webp" height="100rem"/></center>
+
+提问前请仔细查看[项目文档](https://asrt.ailemon.net/docs/)、 
+[FAQ常见问题](https://wiki.ailemon.net/docs/asrt-doc/asrt-doc-1deoeud494h4f)
+以及[Issues](https://github.com/nl8590687/ASRT_SpeechRecognition/issues) 避免重复提问
+
+如果程序运行时有任何异常情况，在提问时请发出完整截图，并注明所使用的CPU架构，GPU型号，操作系统、Python，TensorFlow和CUDA版本，以及是否修改过任何代码或增删数据集等。
+
+## Introduction 简介
+
+本项目使用tensorFlow.keras基于深度卷积神经网络和长短时记忆神经网络、注意力机制以及CTC实现。
+
+## 训练模型的最低软硬件要求
+### 硬件
+* CPU: 4核 (x86_64, amd64) +
+* RAM: 16 GB +
+* GPU: NVIDIA, Graph Memory 11GB+ (1080ti起步)
+* 硬盘: 500 GB 机械硬盘(或固态硬盘)
+
+### 软件
+* Linux: Ubuntu 18.04 + / CentOS 7 + 或 Windows 10/11
+* Python: 3.7 - 3.10 及后续版本
+* TensorFlow: 2.5 - 2.11 及后续版本
+
+## 快速开始
+
+以在Linux系统下的操作为例：
+
+首先通过Git将本项目克隆到您的计算机上，然后下载本项目训练所需要的数据集，下载链接详见[文档末尾部分](https://github.com/nl8590687/ASRT_SpeechRecognition#data-sets-%E6%95%B0%E6%8D%AE%E9%9B%86)。
+```shell
+$ git clone https://github.com/nl8590687/ASRT_SpeechRecognition.git
+```
+
+或者您也可以通过 "Fork" 按钮，将本项目Copy一份副本，然后通过您自己的SSH密钥克隆到本地。
+
+通过git克隆仓库以后，进入项目根目录；并创建一个存储数据的子目录， 例如 `/data/speech_data` (可使用软链接代替)，然后将下载好的数据集直接解压进去
+
+注意，当前版本中，在配置文件里，默认添加了Thchs30、ST-CMDS、Primewords、aishell-1、aidatatang200、MagicData 六个数据集，如果不需要请自行删除。如果要使用其他数据集需要自行添加数据配置，并提前使用ASRT支持的标准格式整理数据。
+
+```shell
+$ cd ASRT_SpeechRecognition
+
+$ mkdir /data/speech_data
+
+$ tar zxf <数据集压缩文件名> -C /data/speech_data/ 
+```
+
+下载默认数据集的拼音标签文件：
+```shell
+$ python download_default_datalist.py
+```
+
+目前可用的模型有24、25、251和251bn
+
+运行本项目之前，请安装必要的[Python3版依赖库](https://github.com/nl8590687/ASRT_SpeechRecognition#python-import)
+
+本项目开始训练请执行：
+```shell
+$ python3 train_speech_model.py
+```
+本项目开始测试请执行：
+```shell
+$ python3 evaluate_speech_model.py
+```
+测试之前，请确保代码中填写的模型文件路径存在。
+
+预测单条音频文件的语音识别文本：
+```shell
+$ python3 predict_speech_file.py
+```
+
+启动ASRT HTTP协议的API服务器启动请执行：
+```shell
+$ python3 asrserver_http.py
+```
+
+本地测试调用HTTP协议API服务是否成功：
+```shell
+$ python3 client_http.py
+```
+
+启动ASRT GRPC协议的API服务器启动请执行：
+```shell
+$ python3 asrserver_grpc.py
+```
+
+本地测试调用GRPC协议API服务是否成功：
+```shell
+$ python3 client_grpc.py
+```
+
+请注意，开启API服务器之后，需要使用本ASRT项目对应的客户端软件来进行语音识别，详见Wiki文档[下载ASRT语音识别客户端SDK和Demo](https://wiki.ailemon.net/docs/asrt-doc/download)。
+
+如果要训练和使用非251bn版模型，请在代码�� `from speech_model.xxx import xxx` 的相应位置做修改。
+
+使用docker直接部署ASRT：
+```shell
+$ docker pull ailemondocker/asrt_service:1.3.0
+$ docker run --rm -it -p 20001:20001 -p 20002:20002 --name asrt-server -d ailemondocker/asrt_service:1.3.0
+```
+仅CPU运行推理识别，不作训练
+
+## Model 模型
+
+### Speech Model 语音模型
+
+DCNN + CTC
+
+其中，输入的音频的最大时间长度为16秒，输出为对应的汉语拼音序列
+
+* 关于下载已经训练好的模型的问题
+
+已经训练好的模型包含在发布版服务端程序压缩包里面，发布版成品服务端程序可以在此下载：[ASRT下载页面](https://wiki.ailemon.net/docs/asrt-doc/download)。
+
+Github本仓库下[Releases](https://github.com/nl8590687/ASRT_SpeechRecognition/releases)页面里面还包括各个不同版本的介绍信息，每个版本下方的zip压缩包也是包含已经训练好的模型的发布版服务端程序压缩包。
+
+### Language Model 语言模型
+
+基于概率图的最大熵隐马尔可夫模型
+
+输入为汉语拼音序列，输出为对应的汉字文本
+
+## About Accuracy 关于准确率
+
+当前，最好的模型在测试集上基本能达到85%的汉语拼音正确率
+
+## Python依赖库
+
+* tensorFlow (2.5-2.11+)
+* numpy
+* wave
+* matplotlib
+* scipy
+* requests
+* flask
+* waitress
+* grpcio / grpcio-tools / protobuf
+
+不会安装环境的同学请直接运行以下命令(前提是有GPU且已经安装好 Python3.9、CUDA 11.2 和 cudnn 8.1)：
+
+```shell
+$ pip install -r requirements.txt
+```
+
+[依赖环境和性能配置要求](https://wiki.ailemon.net/docs/asrt-doc/asrt-doc-1deobk7bmlgd6)
+
+## Data Sets 数据集
+
+完整内容请查看：[几个最新免费开源的中文语音数据集](https://blog.ailemon.net/2018/11/21/free-open-source-chinese-speech-datasets/)
+
+|数据集|时长|大小|国内下载|国外下载|
+|-|-|-|-|-|
+|THCHS30|40h|6.01G|[data_thchs30.tgz](<http://openslr.magicdatatech.com/resources/18/data_thchs30.tgz>)|[data_thchs30.tgz](<http://www.openslr.org/resources/18/data_thchs30.tgz>)|
+|ST-CMDS|100h|7.67G|[ST-CMDS-20170001_1-OS.tar.gz](<http://openslr.magicdatatech.com/resources/38/ST-CMDS-20170001_1-OS.tar.gz>)|[ST-CMDS-20170001_1-OS.tar.gz](<http://www.openslr.org/resources/38/ST-CMDS-20170001_1-OS.tar.gz>)|
+|AIShell-1|178h|14.51G|[data_aishell.tgz](<http://openslr.magicdatatech.com/resources/33/data_aishell.tgz>)|[data_aishell.tgz](<http://www.openslr.org/resources/33/data_aishell.tgz>)|
+|Primewords|100h|8.44G|[primewords_md_2018_set1.tar.gz](<http://openslr.magicdatatech.com/resources/47/primewords_md_2018_set1.tar.gz>)|[primewords_md_2018_set1.tar.gz](<http://www.openslr.org/resources/47/primewords_md_2018_set1.tar.gz>)|
+|MagicData|755h|52G/1.0G/2.2G| [train_set.tar.gz](<http://openslr.magicdatatech.com/resources/68/train_set.tar.gz>) / [dev_set.tar.gz](<http://openslr.magicdatatech.com/resources/68/dev_set.tar.gz>) / [test_set.tar.gz](<http://openslr.magicdatatech.com/resources/68/test_set.tar.gz>)|[train_set.tar.gz](<http://www.openslr.org/resources/68/train_set.tar.gz>) / [dev_set.tar.gz](<http://www.openslr.org/resources/68/dev_set.tar.gz>) / [test_set.tar.gz](<http://www.openslr.org/resources/68/test_set.tar.gz>)|
+
+  注：AISHELL-1 数据集解压方法
+
+  ```
+  $ tar xzf data_aishell.tgz
+  $ cd data_aishell/wav
+  $ for tar in *.tar.gz;  do tar xvf $tar; done
+  ```
+
+特别鸣谢！感谢前辈们的公开语音数据集
+
+如果提供的数据集链接无法打开和下载，请点击该链接 [OpenSLR](http://www.openslr.org)
+
+## ASRT语音识别API客户端调用SDK
+
+ASRT为客户端通过RPC方式调用开发语音识别功能提供了不同平台和编程语言的SDK接入能力，对于其他平台，可直接通过调用通用RESTful Open API方式进行语音识别功能接入。具体接入步骤请看ASRT项目文档。
+
+|客户端平台|项目仓库链接|
+|-|-|
+|Windows客户端SDK和Demo|[ASRT_SDK_WinClient](https://github.com/nl8590687/ASRT_SDK_WinClient)|
+|跨平台Python3客户端SDK和Demo|[ASRT_SDK_Python3](https://github.com/nl8590687/ASRT_SDK_Python3)|
+|跨平台Golang客户端SDK和Demo|[asrt-sdk-go](https://github.com/nl8590687/asrt-sdk-go)|
+|Java客户端SDK和Demo|[ASRT_SDK_Java](https://github.com/nl8590687/ASRT_SDK_Java)|
+
+## ASRT相关资料 
+* [查看ASRT项目的Wiki文档](https://wiki.ailemon.net/docs/asrt-doc)
+
+ASRT的原理请查看本文：
+* [ASRT：一个中文语音识别系统](https://blog.ailemon.net/2018/08/29/asrt-a-chinese-speech-recognition-system/)
+
+ASRT训练和部署教程请看：
+* [教你如何使用ASRT训练中文语音识别模型](<https://blog.ailemon.net/2020/08/20/teach-you-how-use-asrt-train-chinese-asr-model/>)
+* [教你如何使用ASRT部署中文语音识别API服务器](<https://blog.ailemon.net/2020/08/27/teach-you-how-use-asrt-deploy-chinese-asr-api-server/>)
+
+关于经常被问到的统计语言模型原理的问题，请看：
+
+* [统计语言模型：从中文拼音到文本](https://blog.ailemon.net/2017/04/27/statistical-language-model-chinese-pinyin-to-words/)
+* [统计N元语言模型生成算法：简单中文词频统计](https://blog.ailemon.net/2017/02/20/simple-words-frequency-statistic-without-segmentation-algorithm/)
+
+关于CTC的问题请看：
+
+* [[翻译]使用CTC进行序列建模](<https://blog.ailemon.net/2019/07/18/sequence-modeling-with-ctc/>)
+
+更多内容请访问作者的博客：[AI柠檬博客](https://blog.ailemon.net/)
+
+或使用[AI柠檬站内搜索引擎](https://s.ailemon.net/)进行相关信息的搜索
+
+## License 开源许可协议
+
+[GPL v3.0](LICENSE) © [nl8590687](https://github.com/nl8590687) 作者：[AI柠檬](https://www.ailemon.net/)
+
+## 参考引用本项目
+
+[DOI: 10.5281/zenodo.5808434](https://doi.org/10.5281/zenodo.5808434)
+
+## Contributors 贡献者们
+
+[贡献者页面](https://github.com/nl8590687/ASRT_SpeechRecognition/graphs/contributors)
+
+@nl8590687 (repo owner)

README_EN.md

@@ -0,0 +1,235 @@
+![](assets/asrt_title_header_en.png)
+
+[![GPL-3.0 Licensed](https://img.shields.io/badge/License-GPL3.0-blue.svg?style=flat)](https://opensource.org/licenses/GPL-3.0) 
+[![Stars](https://img.shields.io/github/stars/nl8590687/ASRT_SpeechRecognition)](https://github.com/nl8590687/ASRT_SpeechRecognition) 
+[![TensorFlow Version](https://img.shields.io/badge/Tensorflow-2.5+-blue.svg)](https://www.tensorflow.org/) 
+[![Python Version](https://img.shields.io/badge/Python-3.7+-blue.svg)](https://www.python.org/) 
+[![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.5808434.svg)](https://doi.org/10.5281/zenodo.5808434)
+
+ASRT is A Deep-Learning-Based Chinese Speech Recognition System. If you like this project, please **star** it. 
+
+**ReadMe Language** | [中文版](https://github.com/nl8590687/ASRT_SpeechRecognition/blob/master/README.md) | English |
+
+[**ASRT Project Home Page**](https://asrt.ailemon.net/) | 
+[**Released Download**](https://wiki.ailemon.net/docs/asrt-doc/download) | 
+[**View this project's wiki document (Chinese)**](https://wiki.ailemon.net/docs/asrt-doc) | 
+[**Experience Demo**](https://asrt.ailemon.net/demo) | 
+[**Donate**](https://wiki.ailemon.net/docs/asrt-doc/asrt-doc-1deo9u61unti9)
+
+If you have any questions in your works with this project, welcome to put up issues in this repo and I will response as soon as possible. 
+
+You can check the [FAQ Page (Chinese)](https://wiki.ailemon.net/docs/asrt-doc/asrt-doc-1deoeud494h4f) first before asking questions to avoid repeating questions.
+
+If there is any abnormality when the program is running, please send a complete screenshot when asking questions, and indicate the CPU architecture, GPU model, operating system, Python, TensorFlow and CUDA versions used, and whether any code has been modified or data sets have been added or deleted, etc. .
+
+## Introduction
+
+This project uses tensorFlow.keras based on deep convolutional neural network and long-short memory neural network, attention mechanism and CTC to implement. 
+
+## Minimum requirements for training
+### Hardware
+* CPU: 4 Core (x86_64, amd64) +
+* RAM: 16 GB +
+* GPU: NVIDIA, Graph Memory 11GB+ (>1080ti)
+* 硬盘: 500 GB HDD(or SSD)
+
+### Software
+* Linux: Ubuntu 18.04 + / CentOS 7 + or Windows 10/11
+* Python: 3.7 - 3.10 and later
+* TensorFlow: 2.5 - 2.11 and later
+
+## Quick Start
+Take the operation under the Linux system as an example:
+
+First, clone the project to your computer through Git, and then download the data sets needed for the training of this project. For the download links, please refer to [End of Document](https://github.com/nl8590687/ASRT_SpeechRecognition/blob/master/README_EN.md#data-sets)
+```shell
+$ git clone https://github.com/nl8590687/ASRT_SpeechRecognition.git
+```
+
+Or you can use the "Fork" button to copy a copy of the project and then clone it locally with your own SSH key.
+
+After cloning the repository via git, go to the project root directory; create a subdirectory `/data/speech_data` (you can use a soft link instead) for datasets, and then extract the downloaded datasets directly into it.
+
+```shell
+$ cd ASRT_SpeechRecognition
+
+$ mkdir /data/speech_data
+
+$ tar zxf <dataset zip files name> -C /data/speech_data/ 
+```
+
+Note that in the current version, in the configuration file, six data sets, Thchs30, ST-CMDS, Primewords, aishell-1, aidatatang200, MagicData, are added by default, please delete them if you don’t need them. If you want to use other data sets, you need to add data configuration yourself, and use the standard format supported by ASRT to organize the data in advance.
+
+To download pinyin syllable list files for default dataset:
+```shell
+$ python download_default_datalist.py
+```
+
+Currently available models are 24, 25, 251 and 251bn
+
+Before running this project, please install the necessary [Python3 version dependent library](https://github.com/nl8590687/ASRT_SpeechRecognition#python-import)
+
+To start training this project, please execute:
+```shell
+$ python3 train_speech_model.py
+```
+To start the test of this project, please execute:
+```shell
+$ python3 evaluate_speech_model.py
+```
+Before testing, make sure the model file path filled in the code files exists.
+
+To predict one wave audio file for speech recognition：
+```shell
+$ python3 predict_speech_file.py
+```
+
+To startup ASRT API Server with HTTP protocol please execute:
+```shell
+$ python3 asrserver_http.py
+```
+
+Please note that after opening the API server, you need to use the client software corresponding to this ASRT project for voice recognition. For details, see the Wiki documentation to [download ASRT Client SDK & Demo](https://wiki.ailemon.net/docs/asrt-doc/download).
+
+
+To test whether it is successful or not that calls api service interface with HTTP protocol:
+```shell
+$ python3 client_http.py
+```
+
+To startup ASRT API Server with GRPC protocol please execute:
+```shell
+$ python3 asrserver_grpc.py
+```
+
+To test whether it is successful or not that calls api service interface with GRPC protocol:
+```shell
+$ python3 client_grpc.py
+```
+
+If you want to train and use other model(not Model 251bn), make changes in the corresponding position of the `from speech_model.xxx import xxx` in the code files.
+
+If there is any problem during the execution of the program or during use, it can be promptly put forward in the issue, and I will reply as soon as possible.
+
+Deploy ASRT by docker：
+```shell
+$ docker pull ailemondocker/asrt_service:1.3.0
+$ docker run --rm -it -p 20001:20001 -p 20002:20002 --name asrt-server -d ailemondocker/asrt_service:1.3.0
+```
+It will start a api server for recognition rather than training.
+
+## Model
+
+### Speech Model
+
+DCNN + CTC
+
+The maximum length of the input audio is 16 seconds, and the output is the corresponding Chinese pinyin sequence. 
+
+* Questions about downloading trained models
+
+The released finished software that includes trained model weights can be downloaded from [ASRT download page](https://wiki.ailemon.net/docs/asrt-doc/download). 
+
+Github [Releases](https://github.com/nl8590687/ASRT_SpeechRecognition/releases) page includes the archives of the various versions of the software released and it's introduction. Under each version module, there is a zip file that includes trained model weights files. 
+
+### Language Model 
+
+Maximum Entropy Hidden Markov Model Based on Probability Graph. 
+
+The input is a Chinese pinyin sequence, and the output is the corresponding Chinese character text. 
+
+## About Accuracy
+
+At present, the best model can basically reach 85% of Pinyin correct rate on the test set. 
+
+## Python Dependency Library
+
+* tensorFlow (2.5-2.11+)
+* numpy
+* wave
+* matplotlib
+* math
+* scipy
+* requests
+* flask
+* waitress
+* grpcio / grpcio-tools / protobuf
+
+If you have trouble when install those packages, please run the following script to do it as long as you have a GPU and python 3.9, CUDA 11.2 and cudnn 8.1 have been installed：
+
+```shell
+$ pip install -r requirements.txt
+```
+
+[Dependent Environment Details and Hardware Requirement](https://wiki.ailemon.net/docs/asrt-doc/asrt-doc-1deobk7bmlgd6)
+
+## ASRT Client SDK for Calling Speech Recognition API
+
+ASRT provides the abilities to import client SDKs for several platform and programing language for client develop speech recognition features , which work by RPC. Please refer ASRT project documents for detail.
+
+|Client Platform|Project Repos Link|
+|-|-|
+|Windows Client SDK & Demo|[ASRT_SDK_WinClient](https://github.com/nl8590687/ASRT_SDK_WinClient)|
+|Python3 Client SDK & Demo (Any Platform)|[ASRT_SDK_Python3](https://github.com/nl8590687/ASRT_SDK_Python3)|
+|Golang Client SDK & Demo|[asrt-sdk-go](https://github.com/nl8590687/asrt-sdk-go)|
+|Java Client SDK & Demo|[ASRT_SDK_Java](https://github.com/nl8590687/ASRT_SDK_Java)|
+
+## Data Sets 
+
+For full content please refer: [Some free Chinese speech datasets (Chinese)](https://blog.ailemon.net/2018/11/21/free-open-source-chinese-speech-datasets/)
+
+|Dataset|Time|Size|Download (CN Mirrors)|Download (Source)|
+|-|-|-|-|-|
+|THCHS30|40h|6.01G|[data_thchs30.tgz](<http://openslr.magicdatatech.com/resources/18/data_thchs30.tgz>)|[data_thchs30.tgz](<http://www.openslr.org/resources/18/data_thchs30.tgz>)|
+|ST-CMDS|100h|7.67G|[ST-CMDS-20170001_1-OS.tar.gz](<http://openslr.magicdatatech.com/resources/38/ST-CMDS-20170001_1-OS.tar.gz>)|[ST-CMDS-20170001_1-OS.tar.gz](<http://www.openslr.org/resources/38/ST-CMDS-20170001_1-OS.tar.gz>)|
+|AIShell-1|178h|14.51G|[data_aishell.tgz](<http://openslr.magicdatatech.com/resources/33/data_aishell.tgz>)|[data_aishell.tgz](<http://www.openslr.org/resources/33/data_aishell.tgz>)|
+|Primewords|100h|8.44G|[primewords_md_2018_set1.tar.gz](<http://openslr.magicdatatech.com/resources/47/primewords_md_2018_set1.tar.gz>)|[primewords_md_2018_set1.tar.gz](<http://www.openslr.org/resources/47/primewords_md_2018_set1.tar.gz>)|
+|MagicData|755h|52G/1.0G/2.2G| [train_set.tar.gz](<http://openslr.magicdatatech.com/resources/68/train_set.tar.gz>) / [dev_set.tar.gz](<http://openslr.magicdatatech.com/resources/68/dev_set.tar.gz>) / [test_set.tar.gz](<http://openslr.magicdatatech.com/resources/68/test_set.tar.gz>)|[train_set.tar.gz](<http://www.openslr.org/resources/68/train_set.tar.gz>) / [dev_set.tar.gz](<http://www.openslr.org/resources/68/dev_set.tar.gz>) / [test_set.tar.gz](<http://www.openslr.org/resources/68/test_set.tar.gz>)|
+
+
+  Note：The way to unzip AISHELL-1 dataset
+
+  ```
+  $ tar xzf data_aishell.tgz
+  $ cd data_aishell/wav
+  $ for tar in *.tar.gz;  do tar xvf $tar; done
+  ```
+
+Special thanks! Thanks to the predecessors' public voice data set. 
+
+If the provided dataset link cannot be opened and downloaded, click this link [OpenSLR](http://www.openslr.org)
+
+## ASRT Docuemnts
+
+* [ASRT project's Wiki document](https://wiki.ailemon.net/docs/asrt-doc)
+
+A post about ASRT's introduction 
+* [ASRT: Chinese Speech Recognition System (Chinese)](https://blog.ailemon.net/2018/08/29/asrt-a-chinese-speech-recognition-system/)
+
+About how to use ASRT to train and deploy：
+* [Teach you how to use ASRT to train Chinese ASR model (Chinese)](<https://blog.ailemon.net/2020/08/20/teach-you-how-use-asrt-train-chinese-asr-model/>)
+* [Teach you how to use ASRT to deploy Chinese ASR API Server (Chinese)](<https://blog.ailemon.net/2020/08/27/teach-you-how-use-asrt-deploy-chinese-asr-api-server/>)
+
+For questions about the principles of the statistical language model that are often asked, see: 
+* [Simple Chinese word frequency statistics to generate N-gram language model (Chinese)](https://blog.ailemon.net/2017/02/20/simple-words-frequency-statistic-without-segmentation-algorithm/)
+* [Statistical Language Model: Chinese Pinyin to Words (Chinese)](https://blog.ailemon.net/2017/04/27/statistical-language-model-chinese-pinyin-to-words/)
+
+For questions about CTC, see: 
+
+* [[Translation] Sequence Modeling with CTC (Chinese)](<https://blog.ailemon.net/2019/07/18/sequence-modeling-with-ctc/>)
+
+For more infomation please refer to author's blog website: [AILemon Blog](https://blog.ailemon.net/) (Chinese)
+
+## License
+
+[GPL v3.0](LICENSE) © [nl8590687](https://github.com/nl8590687) Author: [ailemon](https://www.ailemon.net/)
+
+## Cite this project
+
+[DOI: 10.5281/zenodo.5808434](https://doi.org/10.5281/zenodo.5808434)
+
+## Contributors
+
+[Contributors Page](https://github.com/nl8590687/ASRT_SpeechRecognition/graphs/contributors)
+
+@nl8590687 (repo owner)

asrserver_grpc.py

@@ -0,0 +1,180 @@
+# !/usr/bin/env python3
+# -*- coding: utf-8 -*-
+#
+# Copyright 2016-2099 Ailemon.net
+#
+# This file is part of ASRT Speech Recognition Tool.
+#
+# ASRT is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+# ASRT is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with ASRT.  If not, see <https://www.gnu.org/licenses/>.
+# ============================================================================
+
+"""
+@author: nl8590687
+ASRT语音识别基于gRPC协议的API服务器程序
+"""
+
+import argparse
+import time
+from concurrent import futures
+import grpc
+
+from assets.asrt_pb2_grpc import AsrtGrpcServiceServicer, add_AsrtGrpcServiceServicer_to_server
+from assets.asrt_pb2 import SpeechResponse, TextResponse
+from speech_model import ModelSpeech
+from model_zoo.speech_model.keras_backend import SpeechModel251BN
+from speech_features import Spectrogram
+from language_model3 import ModelLanguage
+from utils.ops import decode_wav_bytes
+
+API_STATUS_CODE_OK = 200000  # OK
+API_STATUS_CODE_OK_PART = 206000  # 部分结果OK，用于stream
+API_STATUS_CODE_CLIENT_ERROR = 400000
+API_STATUS_CODE_CLIENT_ERROR_FORMAT = 400001  # 请求数据格式错误
+API_STATUS_CODE_CLIENT_ERROR_CONFIG = 400002  # 请求数据配置不支持
+API_STATUS_CODE_SERVER_ERROR = 500000
+API_STATUS_CODE_SERVER_ERROR_RUNNING = 500001  # 服务器运行中出错
+
+parser = argparse.ArgumentParser(description='ASRT gRPC Protocol API Service')
+parser.add_argument('--listen', default='0.0.0.0', type=str, help='the network to listen')
+parser.add_argument('--port', default='20002', type=str, help='the port to listen')
+args = parser.parse_args()
+
+AUDIO_LENGTH = 1600
+AUDIO_FEATURE_LENGTH = 200
+CHANNELS = 1
+# 默认输出的拼音的表示大小是1428，即1427个拼音+1个空白块
+OUTPUT_SIZE = 1428
+sm251bn = SpeechModel251BN(
+    input_shape=(AUDIO_LENGTH, AUDIO_FEATURE_LENGTH, CHANNELS),
+    output_size=OUTPUT_SIZE
+)
+feat = Spectrogram()
+ms = ModelSpeech(sm251bn, feat, max_label_length=64)
+ms.load_model('save_models/' + sm251bn.get_model_name() + '.model.h5')
+
+ml = ModelLanguage('model_language')
+ml.load_model()
+
+_ONE_DAY_IN_SECONDS = 60 * 60 * 24
+
+
+class ApiService(AsrtGrpcServiceServicer):
+    """
+    继承AsrtGrpcServiceServicer,实现hello方法
+    """
+
+    def __init__(self):
+        pass
+
+    def Speech(self, request, context):
+        """
+        具体实现Speech的方法, 并按照pb的返回对象构造SpeechResponse返回
+        :param request:
+        :param context:
+        :return:
+        """
+        wav_data = request.wav_data
+        wav_samples = decode_wav_bytes(samples_data=wav_data.samples,
+                                       channels=wav_data.channels, byte_width=wav_data.byte_width)
+        result = ms.recognize_speech(wav_samples, wav_data.sample_rate)
+        print("语音识别声学模型结果:", result)
+        return SpeechResponse(status_code=API_STATUS_CODE_OK, status_message='',
+                              result_data=result)
+
+    def Language(self, request, context):
+        """
+        具体实现Language的方法, 并按照pb的返回对象构造TextResponse返回
+        :param request:
+        :param context:
+        :return:
+        """
+        print('Language收到了请求:', request)
+        result = ml.pinyin_to_text(list(request.pinyins))
+        print('Language结果:', result)
+        return TextResponse(status_code=API_STATUS_CODE_OK, status_message='',
+                            text_result=result)
+
+    def All(self, request, context):
+        """
+        具体实现All的方法, 并按照pb的返回对象构造TextResponse返回
+        :param request:
+        :param context:
+        :return:
+        """
+        wav_data = request.wav_data
+        wav_samples = decode_wav_bytes(samples_data=wav_data.samples,
+                                       channels=wav_data.channels, byte_width=wav_data.byte_width)
+        result_speech = ms.recognize_speech(wav_samples, wav_data.sample_rate)
+        result = ml.pinyin_to_text(result_speech)
+        print("语音识别结果:", result)
+        return TextResponse(status_code=API_STATUS_CODE_OK, status_message='',
+                            text_result=result)
+
+    def Stream(self, request_iterator, context):
+        """
+        具体实现Stream的方法, 并按照pb的返回对象构造TextResponse返回
+        :param request_iterator:
+        :param context:
+        :return:
+        """
+        result = list()
+        tmp_result_last = list()
+        beam_size = 100
+
+        for request in request_iterator:
+            wav_data = request.wav_data
+            wav_samples = decode_wav_bytes(samples_data=wav_data.samples,
+                                           channels=wav_data.channels,
+                                           byte_width=wav_data.byte_width)
+            result_speech = ms.recognize_speech(wav_samples, wav_data.sample_rate)
+
+            for item_pinyin in result_speech:
+                tmp_result = ml.pinyin_stream_decode(tmp_result_last, item_pinyin, beam_size)
+                if len(tmp_result) == 0 and len(tmp_result_last) > 0:
+                    result.append(tmp_result_last[0][0])
+                    print("流式语音识别结果：", ''.join(result))
+                    yield TextResponse(status_code=API_STATUS_CODE_OK, status_message='',
+                                       text_result=''.join(result))
+                    result = list()
+
+                    tmp_result = ml.pinyin_stream_decode([], item_pinyin, beam_size)
+                tmp_result_last = tmp_result
+                yield TextResponse(status_code=API_STATUS_CODE_OK_PART, status_message='',
+                                   text_result=''.join(tmp_result[0][0]))
+
+        if len(tmp_result_last) > 0:
+            result.append(tmp_result_last[0][0])
+            print("流式语音识别结果：", ''.join(result))
+            yield TextResponse(status_code=API_STATUS_CODE_OK, status_message='',
+                               text_result=''.join(result))
+
+
+def run(host, port):
+    """
+    gRPC API服务启动
+    :return:
+    """
+    server = grpc.server(futures.ThreadPoolExecutor(max_workers=10))
+    add_AsrtGrpcServiceServicer_to_server(ApiService(), server)
+    server.add_insecure_port(''.join([host, ':', port]))
+    server.start()
+    print("start service...")
+    try:
+        while True:
+            time.sleep(_ONE_DAY_IN_SECONDS)
+    except KeyboardInterrupt:
+        server.stop(0)
+
+
+if __name__ == '__main__':
+    run(host=args.listen, port=args.port)

asrserver_http.py

@@ -0,0 +1,189 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+#
+# Copyright 2016-2099 Ailemon.net
+#
+# This file is part of ASRT Speech Recognition Tool.
+#
+# ASRT is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+# ASRT is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with ASRT.  If not, see <https://www.gnu.org/licenses/>.
+# ============================================================================
+
+"""
+@author: nl8590687
+ASRT语音识别基于HTTP协议的API服务器程序
+"""
+
+import argparse
+import base64
+import json
+from flask import Flask, Response, request
+
+from speech_model import ModelSpeech
+from model_zoo.speech_model.keras_backend import SpeechModel251BN
+from speech_features import Spectrogram
+from language_model3 import ModelLanguage
+from utils.ops import decode_wav_bytes
+
+API_STATUS_CODE_OK = 200000  # OK
+API_STATUS_CODE_CLIENT_ERROR = 400000
+API_STATUS_CODE_CLIENT_ERROR_FORMAT = 400001  # 请求数据格式错误
+API_STATUS_CODE_CLIENT_ERROR_CONFIG = 400002  # 请求数据配置不支持
+API_STATUS_CODE_SERVER_ERROR = 500000
+API_STATUS_CODE_SERVER_ERROR_RUNNING = 500001  # 服务器运行中出错
+
+parser = argparse.ArgumentParser(description='ASRT HTTP+Json RESTful API Service')
+parser.add_argument('--listen', default='0.0.0.0', type=str, help='the network to listen')
+parser.add_argument('--port', default='20001', type=str, help='the port to listen')
+args = parser.parse_args()
+
+app = Flask("ASRT API Service")
+
+AUDIO_LENGTH = 1600
+AUDIO_FEATURE_LENGTH = 200
+CHANNELS = 1
+# 默认输出的拼音的表示大小是1428，即1427个拼音+1个空白块
+OUTPUT_SIZE = 1428
+sm251bn = SpeechModel251BN(
+    input_shape=(AUDIO_LENGTH, AUDIO_FEATURE_LENGTH, CHANNELS),
+    output_size=OUTPUT_SIZE
+)
+feat = Spectrogram()
+ms = ModelSpeech(sm251bn, feat, max_label_length=64)
+ms.load_model('save_models/' + sm251bn.get_model_name() + '.model.h5')
+
+ml = ModelLanguage('model_language')
+ml.load_model()
+
+
+class AsrtApiResponse:
+    """
+    ASRT语音识别基于HTTP协议的API接口响应类
+    """
+
+    def __init__(self, status_code, status_message='', result=''):
+        self.status_code = status_code
+        self.status_message = status_message
+        self.result = result
+
+    def to_json(self):
+        """
+        类转json
+        """
+        return json.dumps(self, default=lambda o: o.__dict__, sort_keys=True)
+
+
+# api接口根url:GET
+@app.route('/', methods=["GET"])
+def index_get():
+    """
+    根路径handle GET方法
+    """
+    buffer = ''
+    with open('assets/default.html', 'r', encoding='utf-8') as file_handle:
+        buffer = file_handle.read()
+    return Response(buffer, mimetype='text/html; charset=utf-8')
+
+
+# api接口根url:POST
+@app.route('/', methods=["POST"])
+def index_post():
+    """
+    根路径handle POST方法
+    """
+    json_data = AsrtApiResponse(API_STATUS_CODE_OK, 'ok')
+    buffer = json_data.to_json()
+    return Response(buffer, mimetype='application/json')
+
+
+# 获取分类列表
+@app.route('/<level>', methods=["POST"])
+def recognition_post(level):
+    """
+    其他路径 POST方法
+    """
+    # 读取json文件内容
+    try:
+        if level == 'speech':
+            request_data = request.get_json()
+            samples = request_data['samples']
+            wavdata_bytes = base64.urlsafe_b64decode(bytes(samples, encoding='utf-8'))
+            sample_rate = request_data['sample_rate']
+            channels = request_data['channels']
+            byte_width = request_data['byte_width']
+
+            wavdata = decode_wav_bytes(samples_data=wavdata_bytes,
+                                       channels=channels, byte_width=byte_width)
+            result = ms.recognize_speech(wavdata, sample_rate)
+
+            json_data = AsrtApiResponse(API_STATUS_CODE_OK, 'speech level')
+            json_data.result = result
+            buffer = json_data.to_json()
+            print('output:', buffer)
+            return Response(buffer, mimetype='application/json')
+        elif level == 'language':
+            request_data = request.get_json()
+
+            seq_pinyin = request_data['sequence_pinyin']
+
+            result = ml.pinyin_to_text(seq_pinyin)
+
+            json_data = AsrtApiResponse(API_STATUS_CODE_OK, 'language level')
+            json_data.result = result
+            buffer = json_data.to_json()
+            print('output:', buffer)
+            return Response(buffer, mimetype='application/json')
+        elif level == 'all':
+            request_data = request.get_json()
+
+            samples = request_data['samples']
+            wavdata_bytes = base64.urlsafe_b64decode(samples)
+            sample_rate = request_data['sample_rate']
+            channels = request_data['channels']
+            byte_width = request_data['byte_width']
+
+            wavdata = decode_wav_bytes(samples_data=wavdata_bytes,
+                                       channels=channels, byte_width=byte_width)
+            result_speech = ms.recognize_speech(wavdata, sample_rate)
+            result = ml.pinyin_to_text(result_speech)
+
+            json_data = AsrtApiResponse(API_STATUS_CODE_OK, 'all level')
+            json_data.result = result
+            buffer = json_data.to_json()
+            print('ASRT Result:', result, 'output:', buffer)
+            return Response(buffer, mimetype='application/json')
+        else:
+            request_data = request.get_json()
+            print('input:', request_data)
+            json_data = AsrtApiResponse(API_STATUS_CODE_CLIENT_ERROR, '')
+            buffer = json_data.to_json()
+            print('output:', buffer)
+            return Response(buffer, mimetype='application/json')
+    except Exception as except_general:
+        request_data = request.get_json()
+        # print(request_data['sample_rate'], request_data['channels'],
+        # request_data['byte_width'], len(request_data['samples']),
+        # request_data['samples'][-100:])
+        json_data = AsrtApiResponse(API_STATUS_CODE_SERVER_ERROR, str(except_general))
+        buffer = json_data.to_json()
+        # print("input:", request_data, "\n", "output:", buffer)
+        print("output:", buffer, "error:", except_general)
+        return Response(buffer, mimetype='application/json')
+
+
+if __name__ == '__main__':
+    # for development env
+    # app.run(host='0.0.0.0', port=20001)
+    # for production env
+    import waitress
+
+    waitress.serve(app, host=args.listen, port=args.port)

asrt_config.json

@@ -0,0 +1,32 @@
+{
+    "dict_filename": "dict.txt",
+
+    "dataset":{
+        "train":[
+            {
+                "name": "thchs30_train",
+                "data_list": "datalist/thchs30/train.wav.lst",
+                "data_path": "data/speech_data",
+                "label_list": "datalist/thchs30/train.syllable.txt"
+            }
+        ],
+
+        "dev":[
+            {
+                "name": "thchs30_dev",
+                "data_list": "datalist/thchs30/cv.wav.lst",
+                "data_path": "data/speech_data",
+                "label_list": "datalist/thchs30/cv.syllable.txt"
+            }
+        ],
+
+        "test":[
+            {
+                "name": "thchs30_test",
+                "data_list": "datalist/thchs30/test.wav.lst",
+                "data_path": "data/speech_data",
+                "label_list": "datalist/thchs30/test.syllable.txt"
+            }
+        ]
+    }
+}

assets/asrt.proto

@@ -0,0 +1,55 @@
+/* Copyright 2016-2099 Ailemon.net
+
+This file is part of ASRT Speech Recognition Tool.
+
+ASRT is free software: you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation, either version 3 of the License, or
+(at your option) any later version.
+ASRT is distributed in the hope that it will be useful,
+
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with ASRT.  If not, see <https://www.gnu.org/licenses/>.
+============================================================================ */
+
+syntax = "proto3";
+package asrt;
+
+//定义服务接口
+service AsrtGrpcService {
+    rpc Speech (SpeechRequest) returns (SpeechResponse) {}  //一个服务中可以定义多个接口，也就是多个函数功能
+    rpc Language (LanguageRequest) returns (TextResponse) {}
+    rpc All (SpeechRequest) returns (TextResponse) {}
+    rpc Stream (stream SpeechRequest) returns (stream TextResponse) {}
+}
+
+message SpeechRequest {
+    WavData wav_data = 1;
+}
+
+message SpeechResponse {
+    int32 status_code = 1; 
+    string status_message = 2; 
+    repeated string result_data = 3; // 拼音结果
+}
+
+message LanguageRequest {
+    repeated string pinyins = 1;
+}
+
+message TextResponse {
+    int32 status_code = 1; 
+    string status_message = 2; 
+    string text_result = 3;
+}
+
+message WavData{
+    bytes samples = 1; // wav样本点字节
+    int32 sample_rate = 2; // wav采样率
+    int32 channels = 3; // wav通道数
+    int32 byte_width = 4; // wav样本字节宽度
+}

assets/asrt_pb2.py

@@ -0,0 +1,325 @@
+# -*- coding: utf-8 -*-
+# Generated by the protocol buffer compiler.  DO NOT EDIT!
+# source: asrt.proto
+"""Generated protocol buffer code."""
+from google.protobuf import descriptor as _descriptor
+from google.protobuf import message as _message
+from google.protobuf import reflection as _reflection
+from google.protobuf import symbol_database as _symbol_database
+
+# @@protoc_insertion_point(imports)
+
+_sym_db = _symbol_database.Default()
+
+DESCRIPTOR = _descriptor.FileDescriptor(
+    name='asrt.proto',
+    package='asrt',
+    syntax='proto3',
+    serialized_options=None,
+    create_key=_descriptor._internal_create_key,
+    serialized_pb=b'\n\nasrt.proto\x12\x04\x61srt\"0\n\rSpeechRequest\x12\x1f\n\x08wav_data\x18\x01 \x01(\x0b\x32\r.asrt.WavData\"R\n\x0eSpeechResponse\x12\x13\n\x0bstatus_code\x18\x01 \x01(\x05\x12\x16\n\x0estatus_message\x18\x02 \x01(\t\x12\x13\n\x0bresult_data\x18\x03 \x03(\t\"\"\n\x0fLanguageRequest\x12\x0f\n\x07pinyins\x18\x01 \x03(\t\"P\n\x0cTextResponse\x12\x13\n\x0bstatus_code\x18\x01 \x01(\x05\x12\x16\n\x0estatus_message\x18\x02 \x01(\t\x12\x13\n\x0btext_result\x18\x03 \x01(\t\"U\n\x07WavData\x12\x0f\n\x07samples\x18\x01 \x01(\x0c\x12\x13\n\x0bsample_rate\x18\x02 \x01(\x05\x12\x10\n\x08\x63hannels\x18\x03 \x01(\x05\x12\x12\n\nbyte_width\x18\x04 \x01(\x05\x32\xec\x01\n\x0f\x41srtGrpcService\x12\x35\n\x06Speech\x12\x13.asrt.SpeechRequest\x1a\x14.asrt.SpeechResponse\"\x00\x12\x37\n\x08Language\x12\x15.asrt.LanguageRequest\x1a\x12.asrt.TextResponse\"\x00\x12\x30\n\x03\x41ll\x12\x13.asrt.SpeechRequest\x1a\x12.asrt.TextResponse\"\x00\x12\x37\n\x06Stream\x12\x13.asrt.SpeechRequest\x1a\x12.asrt.TextResponse\"\x00(\x01\x30\x01\x62\x06proto3'
+)
+
+_SPEECHREQUEST = _descriptor.Descriptor(
+    name='SpeechRequest',
+    full_name='asrt.SpeechRequest',
+    filename=None,
+    file=DESCRIPTOR,
+    containing_type=None,
+    create_key=_descriptor._internal_create_key,
+    fields=[
+        _descriptor.FieldDescriptor(
+            name='wav_data', full_name='asrt.SpeechRequest.wav_data', index=0,
+            number=1, type=11, cpp_type=10, label=1,
+            has_default_value=False, default_value=None,
+            message_type=None, enum_type=None, containing_type=None,
+            is_extension=False, extension_scope=None,
+            serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key),
+    ],
+    extensions=[
+    ],
+    nested_types=[],
+    enum_types=[
+    ],
+    serialized_options=None,
+    is_extendable=False,
+    syntax='proto3',
+    extension_ranges=[],
+    oneofs=[
+    ],
+    serialized_start=20,
+    serialized_end=68,
+)
+
+_SPEECHRESPONSE = _descriptor.Descriptor(
+    name='SpeechResponse',
+    full_name='asrt.SpeechResponse',
+    filename=None,
+    file=DESCRIPTOR,
+    containing_type=None,
+    create_key=_descriptor._internal_create_key,
+    fields=[
+        _descriptor.FieldDescriptor(
+            name='status_code', full_name='asrt.SpeechResponse.status_code', index=0,
+            number=1, type=5, cpp_type=1, label=1,
+            has_default_value=False, default_value=0,
+            message_type=None, enum_type=None, containing_type=None,
+            is_extension=False, extension_scope=None,
+            serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key),
+        _descriptor.FieldDescriptor(
+            name='status_message', full_name='asrt.SpeechResponse.status_message', index=1,
+            number=2, type=9, cpp_type=9, label=1,
+            has_default_value=False, default_value=b"".decode('utf-8'),
+            message_type=None, enum_type=None, containing_type=None,
+            is_extension=False, extension_scope=None,
+            serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key),
+        _descriptor.FieldDescriptor(
+            name='result_data', full_name='asrt.SpeechResponse.result_data', index=2,
+            number=3, type=9, cpp_type=9, label=3,
+            has_default_value=False, default_value=[],
+            message_type=None, enum_type=None, containing_type=None,
+            is_extension=False, extension_scope=None,
+            serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key),
+    ],
+    extensions=[
+    ],
+    nested_types=[],
+    enum_types=[
+    ],
+    serialized_options=None,
+    is_extendable=False,
+    syntax='proto3',
+    extension_ranges=[],
+    oneofs=[
+    ],
+    serialized_start=70,
+    serialized_end=152,
+)
+
+_LANGUAGEREQUEST = _descriptor.Descriptor(
+    name='LanguageRequest',
+    full_name='asrt.LanguageRequest',
+    filename=None,
+    file=DESCRIPTOR,
+    containing_type=None,
+    create_key=_descriptor._internal_create_key,
+    fields=[
+        _descriptor.FieldDescriptor(
+            name='pinyins', full_name='asrt.LanguageRequest.pinyins', index=0,
+            number=1, type=9, cpp_type=9, label=3,
+            has_default_value=False, default_value=[],
+            message_type=None, enum_type=None, containing_type=None,
+            is_extension=False, extension_scope=None,
+            serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key),
+    ],
+    extensions=[
+    ],
+    nested_types=[],
+    enum_types=[
+    ],
+    serialized_options=None,
+    is_extendable=False,
+    syntax='proto3',
+    extension_ranges=[],
+    oneofs=[
+    ],
+    serialized_start=154,
+    serialized_end=188,
+)
+
+_TEXTRESPONSE = _descriptor.Descriptor(
+    name='TextResponse',
+    full_name='asrt.TextResponse',
+    filename=None,
+    file=DESCRIPTOR,
+    containing_type=None,
+    create_key=_descriptor._internal_create_key,
+    fields=[
+        _descriptor.FieldDescriptor(
+            name='status_code', full_name='asrt.TextResponse.status_code', index=0,
+            number=1, type=5, cpp_type=1, label=1,
+            has_default_value=False, default_value=0,
+            message_type=None, enum_type=None, containing_type=None,
+            is_extension=False, extension_scope=None,
+            serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key),
+        _descriptor.FieldDescriptor(
+            name='status_message', full_name='asrt.TextResponse.status_message', index=1,
+            number=2, type=9, cpp_type=9, label=1,
+            has_default_value=False, default_value=b"".decode('utf-8'),
+            message_type=None, enum_type=None, containing_type=None,
+            is_extension=False, extension_scope=None,
+            serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key),
+        _descriptor.FieldDescriptor(
+            name='text_result', full_name='asrt.TextResponse.text_result', index=2,
+            number=3, type=9, cpp_type=9, label=1,
+            has_default_value=False, default_value=b"".decode('utf-8'),
+            message_type=None, enum_type=None, containing_type=None,
+            is_extension=False, extension_scope=None,
+            serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key),
+    ],
+    extensions=[
+    ],
+    nested_types=[],
+    enum_types=[
+    ],
+    serialized_options=None,
+    is_extendable=False,
+    syntax='proto3',
+    extension_ranges=[],
+    oneofs=[
+    ],
+    serialized_start=190,
+    serialized_end=270,
+)
+
+_WAVDATA = _descriptor.Descriptor(
+    name='WavData',
+    full_name='asrt.WavData',
+    filename=None,
+    file=DESCRIPTOR,
+    containing_type=None,
+    create_key=_descriptor._internal_create_key,
+    fields=[
+        _descriptor.FieldDescriptor(
+            name='samples', full_name='asrt.WavData.samples', index=0,
+            number=1, type=12, cpp_type=9, label=1,
+            has_default_value=False, default_value=b"",
+            message_type=None, enum_type=None, containing_type=None,
+            is_extension=False, extension_scope=None,
+            serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key),
+        _descriptor.FieldDescriptor(
+            name='sample_rate', full_name='asrt.WavData.sample_rate', index=1,
+            number=2, type=5, cpp_type=1, label=1,
+            has_default_value=False, default_value=0,
+            message_type=None, enum_type=None, containing_type=None,
+            is_extension=False, extension_scope=None,
+            serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key),
+        _descriptor.FieldDescriptor(
+            name='channels', full_name='asrt.WavData.channels', index=2,
+            number=3, type=5, cpp_type=1, label=1,
+            has_default_value=False, default_value=0,
+            message_type=None, enum_type=None, containing_type=None,
+            is_extension=False, extension_scope=None,
+            serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key),
+        _descriptor.FieldDescriptor(
+            name='byte_width', full_name='asrt.WavData.byte_width', index=3,
+            number=4, type=5, cpp_type=1, label=1,
+            has_default_value=False, default_value=0,
+            message_type=None, enum_type=None, containing_type=None,
+            is_extension=False, extension_scope=None,
+            serialized_options=None, file=DESCRIPTOR, create_key=_descriptor._internal_create_key),
+    ],
+    extensions=[
+    ],
+    nested_types=[],
+    enum_types=[
+    ],
+    serialized_options=None,
+    is_extendable=False,
+    syntax='proto3',
+    extension_ranges=[],
+    oneofs=[
+    ],
+    serialized_start=272,
+    serialized_end=357,
+)
+
+_SPEECHREQUEST.fields_by_name['wav_data'].message_type = _WAVDATA
+DESCRIPTOR.message_types_by_name['SpeechRequest'] = _SPEECHREQUEST
+DESCRIPTOR.message_types_by_name['SpeechResponse'] = _SPEECHRESPONSE
+DESCRIPTOR.message_types_by_name['LanguageRequest'] = _LANGUAGEREQUEST
+DESCRIPTOR.message_types_by_name['TextResponse'] = _TEXTRESPONSE
+DESCRIPTOR.message_types_by_name['WavData'] = _WAVDATA
+_sym_db.RegisterFileDescriptor(DESCRIPTOR)
+
+SpeechRequest = _reflection.GeneratedProtocolMessageType('SpeechRequest', (_message.Message,), {
+    'DESCRIPTOR': _SPEECHREQUEST,
+    '__module__': 'asrt_pb2'
+    # @@protoc_insertion_point(class_scope:asrt.SpeechRequest)
+})
+_sym_db.RegisterMessage(SpeechRequest)
+
+SpeechResponse = _reflection.GeneratedProtocolMessageType('SpeechResponse', (_message.Message,), {
+    'DESCRIPTOR': _SPEECHRESPONSE,
+    '__module__': 'asrt_pb2'
+    # @@protoc_insertion_point(class_scope:asrt.SpeechResponse)
+})
+_sym_db.RegisterMessage(SpeechResponse)
+
+LanguageRequest = _reflection.GeneratedProtocolMessageType('LanguageRequest', (_message.Message,), {
+    'DESCRIPTOR': _LANGUAGEREQUEST,
+    '__module__': 'asrt_pb2'
+    # @@protoc_insertion_point(class_scope:asrt.LanguageRequest)
+})
+_sym_db.RegisterMessage(LanguageRequest)
+
+TextResponse = _reflection.GeneratedProtocolMessageType('TextResponse', (_message.Message,), {
+    'DESCRIPTOR': _TEXTRESPONSE,
+    '__module__': 'asrt_pb2'
+    # @@protoc_insertion_point(class_scope:asrt.TextResponse)
+})
+_sym_db.RegisterMessage(TextResponse)
+
+WavData = _reflection.GeneratedProtocolMessageType('WavData', (_message.Message,), {
+    'DESCRIPTOR': _WAVDATA,
+    '__module__': 'asrt_pb2'
+    # @@protoc_insertion_point(class_scope:asrt.WavData)
+})
+_sym_db.RegisterMessage(WavData)
+
+_ASRTGRPCSERVICE = _descriptor.ServiceDescriptor(
+    name='AsrtGrpcService',
+    full_name='asrt.AsrtGrpcService',
+    file=DESCRIPTOR,
+    index=0,
+    serialized_options=None,
+    create_key=_descriptor._internal_create_key,
+    serialized_start=360,
+    serialized_end=596,
+    methods=[
+        _descriptor.MethodDescriptor(
+            name='Speech',
+            full_name='asrt.AsrtGrpcService.Speech',
+            index=0,
+            containing_service=None,
+            input_type=_SPEECHREQUEST,
+            output_type=_SPEECHRESPONSE,
+            serialized_options=None,
+            create_key=_descriptor._internal_create_key,
+        ),
+        _descriptor.MethodDescriptor(
+            name='Language',
+            full_name='asrt.AsrtGrpcService.Language',
+            index=1,
+            containing_service=None,
+            input_type=_LANGUAGEREQUEST,
+            output_type=_TEXTRESPONSE,
+            serialized_options=None,
+            create_key=_descriptor._internal_create_key,
+        ),
+        _descriptor.MethodDescriptor(
+            name='All',
+            full_name='asrt.AsrtGrpcService.All',
+            index=2,
+            containing_service=None,
+            input_type=_SPEECHREQUEST,
+            output_type=_TEXTRESPONSE,
+            serialized_options=None,
+            create_key=_descriptor._internal_create_key,
+        ),
+        _descriptor.MethodDescriptor(
+            name='Stream',
+            full_name='asrt.AsrtGrpcService.Stream',
+            index=3,
+            containing_service=None,
+            input_type=_SPEECHREQUEST,
+            output_type=_TEXTRESPONSE,
+            serialized_options=None,
+            create_key=_descriptor._internal_create_key,
+        ),
+    ])
+_sym_db.RegisterServiceDescriptor(_ASRTGRPCSERVICE)
+
+DESCRIPTOR.services_by_name['AsrtGrpcService'] = _ASRTGRPCSERVICE
+
+# @@protoc_insertion_point(module_scope)

assets/asrt_pb2_grpc.py

@@ -0,0 +1,169 @@
+# Generated by the gRPC Python protocol compiler plugin. DO NOT EDIT!
+"""Client and server classes corresponding to protobuf-defined services."""
+import grpc
+
+import assets.asrt_pb2 as asrt__pb2
+
+
+class AsrtGrpcServiceStub(object):
+    """定义服务接口
+    """
+
+    def __init__(self, channel):
+        """Constructor.
+
+        Args:
+            channel: A grpc.Channel.
+        """
+        self.Speech = channel.unary_unary(
+            '/asrt.AsrtGrpcService/Speech',
+            request_serializer=asrt__pb2.SpeechRequest.SerializeToString,
+            response_deserializer=asrt__pb2.SpeechResponse.FromString,
+        )
+        self.Language = channel.unary_unary(
+            '/asrt.AsrtGrpcService/Language',
+            request_serializer=asrt__pb2.LanguageRequest.SerializeToString,
+            response_deserializer=asrt__pb2.TextResponse.FromString,
+        )
+        self.All = channel.unary_unary(
+            '/asrt.AsrtGrpcService/All',
+            request_serializer=asrt__pb2.SpeechRequest.SerializeToString,
+            response_deserializer=asrt__pb2.TextResponse.FromString,
+        )
+        self.Stream = channel.stream_stream(
+            '/asrt.AsrtGrpcService/Stream',
+            request_serializer=asrt__pb2.SpeechRequest.SerializeToString,
+            response_deserializer=asrt__pb2.TextResponse.FromString,
+        )
+
+
+class AsrtGrpcServiceServicer(object):
+    """定义服务接口
+    """
+
+    def Speech(self, request, context):
+        """Missing associated documentation comment in .proto file."""
+        context.set_code(grpc.StatusCode.UNIMPLEMENTED)
+        context.set_details('Method not implemented!')
+        raise NotImplementedError('Method not implemented!')
+
+    def Language(self, request, context):
+        """Missing associated documentation comment in .proto file."""
+        context.set_code(grpc.StatusCode.UNIMPLEMENTED)
+        context.set_details('Method not implemented!')
+        raise NotImplementedError('Method not implemented!')
+
+    def All(self, request, context):
+        """Missing associated documentation comment in .proto file."""
+        context.set_code(grpc.StatusCode.UNIMPLEMENTED)
+        context.set_details('Method not implemented!')
+        raise NotImplementedError('Method not implemented!')
+
+    def Stream(self, request_iterator, context):
+        """Missing associated documentation comment in .proto file."""
+        context.set_code(grpc.StatusCode.UNIMPLEMENTED)
+        context.set_details('Method not implemented!')
+        raise NotImplementedError('Method not implemented!')
+
+
+def add_AsrtGrpcServiceServicer_to_server(servicer, server):
+    rpc_method_handlers = {
+        'Speech': grpc.unary_unary_rpc_method_handler(
+            servicer.Speech,
+            request_deserializer=asrt__pb2.SpeechRequest.FromString,
+            response_serializer=asrt__pb2.SpeechResponse.SerializeToString,
+        ),
+        'Language': grpc.unary_unary_rpc_method_handler(
+            servicer.Language,
+            request_deserializer=asrt__pb2.LanguageRequest.FromString,
+            response_serializer=asrt__pb2.TextResponse.SerializeToString,
+        ),
+        'All': grpc.unary_unary_rpc_method_handler(
+            servicer.All,
+            request_deserializer=asrt__pb2.SpeechRequest.FromString,
+            response_serializer=asrt__pb2.TextResponse.SerializeToString,
+        ),
+        'Stream': grpc.stream_stream_rpc_method_handler(
+            servicer.Stream,
+            request_deserializer=asrt__pb2.SpeechRequest.FromString,
+            response_serializer=asrt__pb2.TextResponse.SerializeToString,
+        ),
+    }
+    generic_handler = grpc.method_handlers_generic_handler(
+        'asrt.AsrtGrpcService', rpc_method_handlers)
+    server.add_generic_rpc_handlers((generic_handler,))
+
+
+# This class is part of an EXPERIMENTAL API.
+class AsrtGrpcService(object):
+    """定义服务接口
+    """
+
+    @staticmethod
+    def Speech(request,
+               target,
+               options=(),
+               channel_credentials=None,
+               call_credentials=None,
+               insecure=False,
+               compression=None,
+               wait_for_ready=None,
+               timeout=None,
+               metadata=None):
+        return grpc.experimental.unary_unary(request, target, '/asrt.AsrtGrpcService/Speech',
+                                             asrt__pb2.SpeechRequest.SerializeToString,
+                                             asrt__pb2.SpeechResponse.FromString,
+                                             options, channel_credentials,
+                                             insecure, call_credentials, compression, wait_for_ready, timeout, metadata)
+
+    @staticmethod
+    def Language(request,
+                 target,
+                 options=(),
+                 channel_credentials=None,
+                 call_credentials=None,
+                 insecure=False,
+                 compression=None,
+                 wait_for_ready=None,
+                 timeout=None,
+                 metadata=None):
+        return grpc.experimental.unary_unary(request, target, '/asrt.AsrtGrpcService/Language',
+                                             asrt__pb2.LanguageRequest.SerializeToString,
+                                             asrt__pb2.TextResponse.FromString,
+                                             options, channel_credentials,
+                                             insecure, call_credentials, compression, wait_for_ready, timeout, metadata)
+
+    @staticmethod
+    def All(request,
+            target,
+            options=(),
+            channel_credentials=None,
+            call_credentials=None,
+            insecure=False,
+            compression=None,
+            wait_for_ready=None,
+            timeout=None,
+            metadata=None):
+        return grpc.experimental.unary_unary(request, target, '/asrt.AsrtGrpcService/All',
+                                             asrt__pb2.SpeechRequest.SerializeToString,
+                                             asrt__pb2.TextResponse.FromString,
+                                             options, channel_credentials,
+                                             insecure, call_credentials, compression, wait_for_ready, timeout, metadata)
+
+    @staticmethod
+    def Stream(request_iterator,
+               target,
+               options=(),
+               channel_credentials=None,
+               call_credentials=None,
+               insecure=False,
+               compression=None,
+               wait_for_ready=None,
+               timeout=None,
+               metadata=None):
+        return grpc.experimental.stream_stream(request_iterator, target, '/asrt.AsrtGrpcService/Stream',
+                                               asrt__pb2.SpeechRequest.SerializeToString,
+                                               asrt__pb2.TextResponse.FromString,
+                                               options, channel_credentials,
+                                               insecure, call_credentials, compression, wait_for_ready, timeout,
+                                               metadata)

assets/default.html

@@ -0,0 +1,25 @@
+<!DOCTYPE html>
+<html>
+	<head>
+		<meta charset="utf-8">
+		<title>ASRT Speech Recognition API</title>
+		<style>
+		body {
+			width: 35em;
+			margin: 0 auto;
+			font-family: Tahoma, Verdana, Arial, sans-serif;
+    	}
+		</style>
+	</head>
+	<body>
+		<h1>ASRT Speech Recognition API</h1>
+		<h3>framework version: 1.0</h3>
+		<p>If you see this page, the ASRT api server is successfully installed and working. </p>
+		<p>For online documentation and support please refer to <a href="https://asrt.ailemon.net">ASRT Project Page</a>.</p>
+		<p>Please call this web api by post menthod. </p>
+		<em>Thank you for using ASRT.</em>
+		<footer>
+			<p style="text-align: center;">Copyright © <a href="https://www.ailemon.net">ailemon.net</a></p>
+		</footer>
+	</body>
+</html>

client_grpc.py

@@ -0,0 +1,124 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+#
+# Copyright 2016-2099 Ailemon.net
+#
+# This file is part of ASRT Speech Recognition Tool.
+#
+# ASRT is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+# ASRT is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with ASRT.  If not, see <https://www.gnu.org/licenses/>.
+# ============================================================================
+
+"""
+@author: nl8590687
+ASRT语音识别asrserver grpc协议测试专用客户端
+"""
+
+import grpc
+import time
+
+from assets.asrt_pb2_grpc import AsrtGrpcServiceStub
+from assets.asrt_pb2 import SpeechRequest, LanguageRequest, WavData
+
+from utils.ops import read_wav_bytes
+
+
+def run_speech():
+    """
+    请求ASRT服务Speech方法
+    :return:
+    """
+    conn = grpc.insecure_channel('127.0.0.1:20002')
+    client = AsrtGrpcServiceStub(channel=conn)
+
+    wav_bytes, sample_rate, channels, sample_width = read_wav_bytes('assets/A11_0.wav')
+    print('sample_width:', sample_width)
+    wav_data = WavData(samples=wav_bytes, sample_rate=sample_rate,
+                       channels=channels, byte_width=sample_width)
+
+    request = SpeechRequest(wav_data=wav_data)
+    time_stamp0 = time.time()
+    response = client.Speech(request)
+    time_stamp1 = time.time()
+    print('time:', time_stamp1 - time_stamp0, 's')
+    print("received:", response.result_data)
+
+
+def run_lan():
+    """
+    请求ASRT服务Language方法
+    :return:
+    """
+    conn = grpc.insecure_channel('127.0.0.1:20002')
+    client = AsrtGrpcServiceStub(channel=conn)
+    pinyin_data = ['ni3', 'hao3', 'ya5']
+    request = LanguageRequest(pinyins=pinyin_data)
+    time_stamp0 = time.time()
+    response = client.Language(request)
+    time_stamp1 = time.time()
+    print('time:', time_stamp1 - time_stamp0, 's')
+    print("received:", response.text_result)
+
+
+def run_all():
+    """
+    请求ASRT服务All方法
+    :return:
+    """
+    conn = grpc.insecure_channel('127.0.0.1:20002')
+    client = AsrtGrpcServiceStub(channel=conn)
+
+    wav_bytes, sample_rate, channels, sample_width = read_wav_bytes('assets/A11_0.wav')
+    print('sample_width:', sample_width)
+    wav_data = WavData(samples=wav_bytes, sample_rate=sample_rate,
+                       channels=channels, byte_width=sample_width)
+
+    request = SpeechRequest(wav_data=wav_data)
+    time_stamp0 = time.time()
+    response = client.All(request)
+    time_stamp1 = time.time()
+    print("received:", response.text_result)
+    print('time:', time_stamp1 - time_stamp0, 's')
+
+
+def run_stream():
+    """
+    请求ASRT服务Stream方法
+    :return:
+    """
+    conn = grpc.insecure_channel('127.0.0.1:20002')
+    client = AsrtGrpcServiceStub(channel=conn)
+
+    wav_bytes, sample_rate, channels, sample_width = read_wav_bytes('assets/A11_0.wav')
+    print('sample_width:', sample_width)
+    wav_data = WavData(samples=wav_bytes, sample_rate=sample_rate,
+                       channels=channels, byte_width=sample_width)
+
+    # 先制造一些客户端能发送的数据
+    def make_some_data():
+        for _ in range(1):
+            time.sleep(1)
+            yield SpeechRequest(wav_data=wav_data)
+
+    try:
+        status_response = client.Stream(make_some_data())
+        for ret in status_response:
+            print("received:", ret.text_result, " , status:", ret.status_code)
+            time.sleep(0.1)
+    except Exception as any_exception:
+        print(f'err in send_status:{any_exception}')
+        return
+
+
+if __name__ == '__main__':
+    # run_all()
+    run_stream()

client_http.py

@@ -0,0 +1,49 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+#
+# Copyright 2016-2099 Ailemon.net
+#
+# This file is part of ASRT Speech Recognition Tool.
+#
+# ASRT is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+# ASRT is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with ASRT.  If not, see <https://www.gnu.org/licenses/>.
+# ============================================================================
+
+'''
+@author: nl8590687
+ASRT语音识别asrserver http协议测试专用客户端
+'''
+import base64
+import json
+import time
+import requests
+from utils.ops import read_wav_bytes
+
+URL = 'http://127.0.0.1:20001/all'
+
+wav_bytes, sample_rate, channels, sample_width = read_wav_bytes('assets/A11_0.wav')
+datas = {
+    'channels': channels,
+    'sample_rate': sample_rate,
+    'byte_width': sample_width,
+    'samples': str(base64.urlsafe_b64encode(wav_bytes), encoding='utf-8')
+}
+headers = {'Content-Type': 'application/json'}
+
+t0=time.time()
+r = requests.post(URL, headers=headers, data=json.dumps(datas))
+t1=time.time()
+r.encoding='utf-8'
+
+result = json.loads(r.text)
+print(result)
+print('time:', t1-t0, 's')

data_loader.py

@@ -0,0 +1,102 @@
+# !/usr/bin/env python3
+# -*- coding: utf-8 -*-
+#
+# Copyright 2016-2099 Ailemon.net
+#
+# This file is part of ASRT Speech Recognition Tool.
+#
+# ASRT is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+# ASRT is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with ASRT.  If not, see <https://www.gnu.org/licenses/>.
+# ============================================================================
+
+"""
+@author: nl8590687
+加载语音数据集用的数据加载器的定义
+"""
+import os
+import random
+import numpy as np
+from utils.config import load_config_file, DEFAULT_CONFIG_FILENAME, load_pinyin_dict
+from utils.ops import read_wav_data
+
+
+class DataLoader:
+    """
+    数据加载器
+
+    参数：\\
+        config: 配置信息字典
+        dataset_type: 要加载的数据集类型，包含('train', 'dev', 'test')三种
+    """
+    def __init__(self, dataset_type:str):
+        self.dataset_type = dataset_type
+
+        self.data_list = list()
+        self.wav_dict = dict()
+        self.label_dict = dict()
+        self.pinyin_list = list()
+        self.pinyin_dict = dict()
+        self._load_data()
+
+    def _load_data(self):
+        config = load_config_file(DEFAULT_CONFIG_FILENAME)
+
+        self.pinyin_list, self.pinyin_dict = load_pinyin_dict(config['dict_filename'])
+
+        for index in range(len(config['dataset'][self.dataset_type])):
+            filename_datalist = config['dataset'][self.dataset_type][index]['data_list']
+            filename_datapath = config['dataset'][self.dataset_type][index]['data_path']
+            with open(filename_datalist, 'r', encoding='utf-8') as file_pointer:
+                lines = file_pointer.read().split('\n')
+                for line in lines:
+                    if len(line) == 0:
+                        continue
+                    tokens = line.split(' ')
+                    self.data_list.append(tokens[0])
+                    self.wav_dict[tokens[0]] = os.path.join(filename_datapath, tokens[1])
+
+            filename_labellist = config['dataset'][self.dataset_type][index]['label_list']
+            with open(filename_labellist, 'r', encoding='utf-8') as file_pointer:
+                lines = file_pointer.read().split('\n')
+                for line in lines:
+                    if len(line) == 0:
+                        continue
+                    tokens = line.split(' ')
+                    self.label_dict[tokens[0]] = tokens[1:]
+
+    def get_data_count(self) -> int:
+        """
+        获取数据集总数量
+        """
+        return len(self.data_list)
+
+    def get_data(self, index: int) -> tuple:
+        """
+        按下标获取一条数据
+        """
+        mark = self.data_list[index]
+
+        wav_signal, sample_rate, _, _ = read_wav_data(self.wav_dict[mark])
+        labels = list()
+        for item in self.label_dict[mark]:
+            if len(item) == 0:
+                continue
+            labels.append(self.pinyin_dict[item])
+
+        data_label = np.array(labels)
+        return wav_signal, sample_rate, data_label
+
+    def shuffle(self) -> None:
+        """
+        随机打乱数据
+        """
+        random.shuffle(self.data_list)

datalist/st-cmds/dev.syllable.txt

@@ -0,0 +1,600 @@
+20170001P00142A0070 mei2 qian2 yi1 fen1 qian2 mei2 you3 lao3 yue1 han4 ka3 li3
+20170001P00196A0072 jiu4 shi4 gan3 jue2 la5 jin1 tian1 he2 xue3 qin2 gei3 wo3 da3 dian4 hua4
+20170001P00337A0004 gao1 zhong1 er4 ban1 du2 ning2 da4 de5 xue1 rui4 qiu1
+20170001P00148I0088 piao1 xue3 ni3 zen3 me5 tui4 chu1 qun2 le5 ne5
+20170001P00131I0047 ling4 wai4 jiu4 jie1 ge4 wen3 shen2 me5 de5 mo1 mo1 er2 yi3
+20170001P00368I0013 wo3 gei3 ni3 wen4 wen4 yun2 xia2 ba5
+20170001P00367A0059 jin1 yin2 cai2 bao3 san1 ge4 ye2 hai2 you3 ge4 di4 di5 jiao4 jun1 bao3
+20170001P00213A0115 na4 ni2 yu4 lin2 shou3 ji1 duo1 shao3 wen4 wen4 ni3 shi4 zai4 shua3 wo3
+20170001P00035I0069 ni3 ken3 ding4 jiu4 shi4 fu4 weng1 le5
+20170001P00267I0078 ba1 dian3 ban4 qi3 chuang2 shua1 ya2 xi3 lian3
+20170001P00344I0013 wo3 ling4 yi2 ge4 qun2 gei3 de5 zhi2 bo1 di4 zhi3
+20170001P00027I0050 hai2 juan1 qian2 gei3 long2 ya3 ren2
+20170001P00090A0025 dai4 ni3 qu4 ri4 yue4 guang1 dou1 yi1 quan1
+20170001P00262A0029 bao3 ju4 qi2 shi4 jing1 shen2 bu4 yun3 xu3 wo3 hou4 tui4 o2
+20170001P00316A0080 ni3 kan4 na4 li3 mian4 you3 lan2 ya2 er3 ji1 ma5
+20170001P00206A0047 mei2 zen3 me5 jiu4 shi4 wen4 yi1 xia4 hao3 hao3 shang4 ke4 sha3 fei2 zhu1
+20170001P00350A0027 da4 er2 zi5 yong3 yuan3 dou1 zai4 tiao3 bo1
+20170001P00325I0013 ni3 zi1 xun2 de5 shi4 qing1 hua1 ci2 chen4 shan1 ma5
+20170001P00071A0011 shuang1 shi2 yi1 zhi1 hou4 fan3 zheng4 shi4 ge4 zhong3 qiong2 o1
+20170001P00091I0112 ping2 gai4 xu1 yao4 de5 hua4 mian4 shi4 xiao3 yuan2 quan1 nei4 de5 hua4 mian4
+20170001P00091I0117 hao3 huai2 nian4 yi3 qian2 neng2 da3 qiu2 de5 ri4 zi5 a5
+20170001P00353I0057 wo3 zhe4 zhou1 wu3 huo4 zhe3 zhou1 liu4 jiu4 zou3 le5
+20170001P00085I0047 gei3 ni3 kan4 kan4 qing1 yun2 xiong2
+20170001P00020I0017 ta1 za3 lian2 zhe4 ge4 dou1 dao4 chu4 xuan1 yang2 ta1 shen2 me5 yi4 si1
+20170001P00120I0087 qu4 zhe4 jiang1 yong3 kang1 diao4 ji1 shou1 dao4 huo4 kuan3
+20170001P00079I0087 wo3 qu4 zhao3 shi1 xiong1 gai4 de5
+20170001P00329I0114 jia1 shang4 weng1 de2 li4 ta1 shi4 bu2 shi4 mei2 zai4 xian4
+20170001P00078A0003 tao2 yuan1 ming2 na4 pian4 yuan4 zi5
+20170001P00189A0089 huang2 ya3 xuan1 ta1 dan1 shen1 ma5 wen4 wo3 gao4 su4 ni3 qu4 wen4
+20170001P00199A0096 ke3 yi3 kan1 cheng1 pi3 pei4 zhang3 sheng4 jiang1 jun1 sha2 dou1 bu2 hui4 le5
+20170001P00183A0046 fu4 gu3 hei1 xue3 fang3 zhang3 qun2 liang3 se4
+20170001P00088A0044 bai3 wan4 fu4 weng1 na4 bian1 hui4 sheng1 zhi2 de5
+20170001P00049A0027 er2 qie3 shang4 ci4 wo3 kan4 dao4 ni3 ma1 ma1 jue2 de5 ta1 hao3 xiong1
+20170001P00065I0104 ri4 qian2 shang4 bao4 qu1 min2 zheng4 ju2 ci2 shan4 hui4
+20170001P00351A0105 ni3 qi1 pian4 gao1 yi1 jun1 gan3 qing2 a5 chun1 meng4 gan4 ma5 a5
+20170001P00032A0033 wo3 qu4 yi1 yuan4 zuo4 he2 ci2 gong4 zhen4
+20170001P00117A0041 bu4 xiang3 gen1 qun2 li3 xiong1 di4 tan2 qian2
+20170001P00256A0119 hai2 pian4 ren2 shuo1 shen2 me5 du4 min2 jun1 jiao4 shou4 shen2 me5 de5
+20170001P00332A0085 ting1 shuo1 ni3 xiang3 guo4 lai2 shua3 shen2 me5 shi2 hou4 lai2
+20170001P00378I0018 wo3 zhe4 ren2 ke3 neng2 jiu4 shi4 tian1 sheng1 re3 ren2 tao3 yan4 de5 ren2
+20170001P00363A0098 dou1 zhi1 dao4 gang1 hua4 jia1 jiao1 bo1 li5 de5 biao1 zhun3
+20170001P00135A0017 qing3 ge4 wei4 quan1 zhang3 zheng3 li3 zi1 liao4 zhun3 shi2 can1 jia1
+20170001P00384I0058 qin1 jie3 yong3 yuan3 kuai4 le4 yong3 yuan3 er4 shi2 ba1 sui4
+20170001P00114A0072 ming2 ming2 sheng1 qi4 a5 er2 qie3 hai2 xiong1 le5
+20170001P00119A0034 shi4 tai4 que1 fa2 yun4 dong4 le5 ni3 sheng1 ri4
+20170001P00148I0096 na4 pian1 ri4 zhi4 shi4 xie3 gei3 ma3 jun1 yu3 de5
+20170001P00163I0041 ni3 yuan4 yi4 yao1 yue1 yi2 ge4 nan2 de5 guo4 lai2
+20170001P00005I0116 zhun3 bu4 zhun3 bei4 huo2 dong4 dou1 gen1 zhao4 shi1 xiong1 hui2 ge4 hua4
+20170001P00349A0065 zai4 qun2 nei4 jin4 zhi3 xuan1 chuan2
+20170001P00316I0061 dou1 bu4 neng2 zuo4 xue2 xiao4 you2 yong3 ka3 le5
+20170001P00007A0092 er2 qie3 hai2 shi4 ge4 gua3 fu4
+20170001P00036I0112 wo3 hai2 shi4 xi3 huan1 pi2 ka3 qiu1
+20170001P00172I0102 qu4 ri4 ben3 liu1 da2 quan1 qu4 ba5
+20170001P00377I0038 rao2 shi4 sha1 man4 di4 guo2 wen2 ming2 xia2 er3 de5 zhong4 xun2 yang2 jian4
+20170001P00347A0048 jian3 fei2 li3 ying2 jun1 xun4 jiu4 ke3 yi3 le5 wo3 wei4 shen2 me5
+20170001P00117A0022 ni3 men5 zhun3 bei4 wan2 er2 dao4 ji3 dian3
+20170001P00269A0096 peng2 you3 ming2 zi4 fan4 jun1 yan4 gao1 er4 wo3 shuo1
+20170001P00371I0083 san1 shan1 qu1 wei3 xuan1 chuan2 bu4
+20170001P00200I0062 ni3 za3 re3 wo3 jia1 lao3 er4 le5 a5
+20170001P00274I0105 wo3 men5 pai4 jian4 er2 huang2 gang1 bu4 bao1 kuo4 er4 ji2 zhan4 dian3 jiu4
+20170001P00209A0013 mu4 ce4 ni3 huai2 le5 ni3 dou1 kuang2 tu3 bu4 zhi3 le5 qin1 ai4 de5
+20170001P00285I0030 qin1 shi4 shang4 shen1 gen1 xia4 shen1 jun1 yun2 de5 shi4 ma5
+20170001P00018A0052 ni3 zhe4 bu4 shi4 yuan1 wang3 hao3 ren2 ne5 ma5
+20170001P00254I0087 wo3 jiu4 guai1 dian3 gun3 dan4 ma5
+20170001P00334I0074 jie2 guo3 yi1 qun2 ren2 du3 zai4 guai3 jiao3 chu4
+20170001P00105I0055 ri4 zhao4 zi3 yu4 xuan1 dian4 hua4
+20170001P00057I0093 bang1 wo3 yue1 zou1 jia1 yun2 he2 wo3 qu4 ya5
+20170001P00306I0069 ni3 shen2 me5 shi2 hou4 ke3 yi3 ci2 zhi2 pi1 zhun3 ya5
+20170001P00214A0013 neng2 ba3 bian4 cha1 zhu1 yu4 shao3 yi1 ren2 de5 ge1 ci2 gei3 wo3 ma5
+20170001P00068A0082 ba3 xiang1 tan2 ji3 jia1 yi1 yuan4 mo1 qing1 chu3
+20170001P00263I0068 huang2 jiao4 zhu3 de5 mou3 shou3 ge1 li3 you3 zhe4 ju4 ge1 ci2
+20170001P00251A0083 gao1 shou3 chu2 le5 da3 ma2 jiang1 xin1 you3 ling2 xi1 wo3 bu2 hui4
+20170001P00263I0019 ni3 hai2 bu2 shi4 yao4 gen1 zhe5 wo3 guo4 qiong2 ri4 zi5
+20170001P00124I0029 wo3 zao3 jiu4 bu4 wan2 wai1 wai1 le5 xian4 zai4 wan2 wei1 bo2
+20170001P00428A0075 mei3 nian2 dou1 yao4 jun1 xun4 jin1 nian2 dou1 di4 er4 nian2 le5
+20170001P00430A0042 mei3 du4 de5 qu4 le5 ai1 kang1 jia1 ye3 shi4
+20170001P00431A0078 dao4 du1 jiang1 yan4 tai4 gui4 xue2 sheng1 xiao1 fei4 bu4 qi3
+20170001P00360A0015 ran2 hou4 ta1 jiu4 shuo1 yao4 ci2 zhi2 ran2 hou4 wo3 men5 jiu4 quan4 ta1
+20170001P00430A0040 zai4 huang2 hua2 you3 ge4 an4 zi5 qian2 ji3 tian1 gang1 hui2 lai2
+20170001P00332I0105 ni3 jin4 la5 xiong2 yue4 de5 qun2 bo1
+20170001P00274A0103 wan2 er2 de5 shen2 me5 yi1 tian1 tian1 de5 yang2 yan2 jun1 zai4 na3 ne5
+20170001P00085I0085 wo3 kan4 dao4 le5 xin1 ji4 sheng1 ren2 tan2 xue3 tan4 xun2
+20170001P00383I0017 qian2 liang3 tian1 hai2 ting1 bie2 ren2 shuo1 ni3 zai4 yun2 nan2
+20170001P00204I0117 ni3 men5 zhun3 bei4 zai4 na3 li3 wan2 ne5
+20170001P00222A0016 ni3 shuo1 nei4 xie1 wa2 wa2 shi4 ni3 he2 qun2 li3 de5 nv3 de5 de5
+20170001P00395I0117 ni3 men5 dian4 you3 xue3 fang3 hei1 duan3 qun2 ma5
+20170001P00430A0102 bu4 neng2 yun3 xu3 fan3 hui3 ma5
+20170001P00368I0039 ni3 dou1 yuan4 yi4 le5 wo3 she3 ming4 pei2 jun1 zi5 ei2
+20170001P00088I0016 huang2 jian4 xiang2 hai2 hui4 kua1 xi1 ban1 ya2 ya5
+20170001P00352I0024 lv3 you2 zhe3 ping2 jun1 zai4 chuan1 dou4 liu2 tian1 shu4
+20170001P00143A0040 gang1 ci2 zhi2 shi4 a5 tian1 cheng2 guang3 gao4 zuo4 guang3 gao4 le5
+20170001P00141A0107 mai3 ying1 xiong2 wo3 hao4 li3 hao3 xiang4 hai2 you3 dian3 quan4
+20170001P00043A0085 ru2 guo3 huang2 jun1 zhen1 de5 zhi3 shi4 wo3 xiang3 wo3 xiang3 de5 zhe4 yang4
+20170001P00035I0105 wo3 zai4 kan4 wo3 men5 jun1 xun4 qun2 li3 liao2 tian1
+20170001P00325I0022 mai3 le5 hao3 duo1 shu1 wo3 ka3 li3 jiu4 qiong2 le5
+20170001P00007A0080 wen4 ti2 jiu4 zai4 yu2 zhang1 shu4 jun1 zhe4 song3 mu4 de2 hen3
+20170001P00406A0033 shi4 yin1 wei4 yang2 xiong2 cai2 che3 qi3 de5
+20170001P00062A0108 er2 qie3 wo3 hui2 da2 wen4 ti2 ye3 xiang4 zhou1 yi2 jun1 na4 yang4
+20170001P00280A0064 qu4 ba5 fu2 hai3 wu2 ya2 lao3 na4 fa3 hao4 li3 qing1
+20170001P00237I0106 wang2 yun2 xia2 gong1 hui4 zhu3 xi2
+20170001P00217A0109 ni3 ge4 diao3 si1 xie3 de5 hao3 wen2 yi4 yu2 ben3 lai2 hui4 you2 yong3
+20170001P00165A0106 zhe4 zhong3 long2 wo3 ji4 de5 you2 xia2 hao4 shang4 ye3 you3
+20170001P00030A0109 li3 yue4 jiu4 zai4 xuan1 xuan1 de5 qun2 le5
+20170001P00091A0043 na4 wo3 jiu4 yong4 na4 ge4 yun3 shi2 hai2 de5
+20170001P00032A0012 wo3 yuan4 de5 te4 se4 zheng3 xing2 xiang4 mu4 jiu4 shi4 zhi1 fang2 long2 xiong1
+20170001P00354I0108 wo3 yao4 shang4 lai2 gai4 ge4 zhang1
+20170001P00107I0002 tu2 shu1 guan3 zhi3 sheng4 xia4 wo3 men5 xue2 yuan4 de5 wa2 er2 le5
+20170001P00192I0062 ri4 ben3 you3 ren2 song4 xue3 zhang4
+20170001P00277A0001 na4 shi4 wo3 de5 zhuan1 you3 ming2 ci2 mao2 xian4 ni3 wen4
+20170001P00389I0103 xin1 xiang1 san1 quan1 yi1 xue2 yuan4 quan2
+20170001P00354I0041 ni3 qu4 hai2 shi4 huang2 xiong1 qu4
+20170001P00324I0108 dou1 shi4 che3 dan4 wan2 wan2 er2 jiu4 shuai3
+20170001P00116A0021 huai2 yu4 ba3 shou3 zhi3 yong4 dao1 zhe2 le5 xian4 zai4 zhu4 yuan4 ne5
+20170001P00365I0019 ni3 chuan1 duan3 qun2 di4 xiong1 wo3 jiu4 chuan1
+20170001P00175A0093 bie2 de5 ren2 mo1 ni3 xiong1 de5 shi2 hou4
+20170001P00166I0071 ni3 zhun3 bei4 ba3 ta1 men5 shuai3 dao4 yi1 bian1 zi4 ji3 wan2 a5
+20170001P00025A0004 ni3 pei2 wo3 you2 shan1 wan2 shui3 qu4 da4 shi1 xiong1 ni3 zen3 me5 le5
+20170001P00396A0077 wo3 gei3 ni3 shi2 jian1 ye3 yun3 xu3 ni3 nan2 guo4
+20170001P00073A0026 ke3 chuan1 sha1 qun2 hao3 you2 yong3 ma5
+20170001P00343I0112 wo3 men5 zai4 si4 hui4 ci2 yun2 si4
+20170001P00352A0067 bao3 bei4 zhe4 xie1 ci2 yong4 de5 shou3 dao4 qin2 lai2
+20170001P00244I0108 ren2 de5 xuan3 ze2 yong3 yuan3 dou1 nan2 yi3 que4 ding4
+20170001P00317A0098 ni3 pei2 xie4 jia1 jiu4 ke3 yi3 la5 xie4 xie4 ge1
+20170001P00109A0022 ran2 hou4 zai4 rang4 lao3 gong1 wen4 yun3 bu4 yun3 xu3
+20170001P00315A0104 ni3 bie2 zai4 hu1 you1 qun2 li3 de5 xiong1 di4 le5
+20170001P00031I0018 jiu4 shi4 gong1 si1 zai4 zhe4 fang1 mian4 que1 fa2 xiang1 guan1 de5 biao1 zhun3
+20170001P00050I0111 wo3 jue2 de5 si4 yuan4 de5 da2 bian4 ci2 man2 you3 yi4 si1 de5
+20170001P00249A0002 er4 wa2 zhu4 yi4 yu3 ju4 hai2 mei2 you3 chu1 lai2 hao3 re4 o2
+20170001P00268I0076 bing4 qie3 shang4 mian4 de5 qi4 pao4 ye3 shi4 fei1 chang2 jun1 yun2 de5
+20170001P00234A0020 ta1 men5 qi1 fu4 wo3 wo3 mei2 you3 a5 wo3 yuan1 wang3 a5
+20170001P00205I0040 zai4 wu3 ning2 lu2 ya2 shan1 you2 lan3
+20170001P00262I0089 nv3 wa2 er2 shuo1 yao4 wen1 ya3 dian3
+20170001P00201I0107 kan4 she4 diao1 ying1 xiong2 chuan2 dou1 ka3
+20170001P00018I0056 wo3 kuai4 ao2 guo4 zhu4 yuan4 ni3 wa2 shi4 na3 tian1 lai2 de5
+20170001P00028A0110 qi3 chuang2 ne5 wo3 jin1 tian1 gen1 peng2 you3 yue1 le5
+20170001P00366I0018 wo3 yong4 jia1 qian2 ba3 ka3 huang2 wang3 zhong1 jian1 ban1 le5 ban1
+20170001P00065I0064 san1 tai2 de5 yi1 yuan4 tai4 lan4 le5
+20170001P00332I0053 wan4 ren2 ci4 ri4 jun1 ren2 liu2 liang4
+20170001P00002A0057 bu4 deng3 yu2 ta1 bu4 cun2 zai4 zao3 wan3 de5 shi4 ni3 yong3 yuan3 ba5
+20170001P00038A0001 ni3 yuan4 yi4 zuo4 wo3 de5 mei3 ren2 yu2 yong3 bu4 pi1 tui3 ma5
+20170001P00364I0119 wo3 yi3 wei2 na4 wa2 yao4 he2 ni3 qun2 cai3
+20170001P00239I0067 wo3 men5 bu4 zai4 qun2 wai1 wai1
+20170001P00339I0053 wo3 yao4 kan4 kan4 ni3 hai2 neng2 che3 chu1 shen2 me5 hua4 ti2 a5
+20170001P00323I0116 ni3 dong3 de5 ba5 zui4 jia1 sun3 you3 wo3 hen3 xi3 huan1 ting1 ne5
+20170001P00144A0003 ni3 zhe4 li3 you3 qu4 nian2 yi1 ri4 juan1 de5 ge4 ren2 ji4 lu4 ma5
+20170001P00353I0052 wo3 zhi3 hui4 wan2 er2 xuan4 wu3 he2 mo2 er3 zhuang1 yuan2
+20170001P00010A0034 bo1 bo1 xuan1 xuan1 zhi4 jing4
+20170001P00227I0072 weng1 niu2 te4 qi2 ren2 min2 fa3 yuan4
+20170001P00397I0085 wen1 zong3 hui2 lai2 le5 yao4 qu4 zhao3 ta1 ma5
+20170001P00095A0016 xu2 xian1 jun1 shen1 qing3 de5 xie2 yi4 fang1 an4 kai1 guo4 le5
+20170001P00360I0081 ta1 ai4 ren2 chen2 jin1 xia2 jie1 guan3 le5 guo2 jin1 zheng4 quan4
+20170001P00386A0078 hun2 shui3 mo2 yu2 kan4 ta1 an1 pai2 da4 yue1 dai4 piao1 liu2 fei4 yong4
+20170001P00409A0114 hai3 di3 lao1 chi1 huo3 guo1 bu2 shi4 hen3 re4
+20170001P00191I0022 gang1 cai2 xiong2 gang1 de5 nv3 peng2 you3 shuo1 huang2 dong1 dong1 hui2 lai2 le5
+20170001P00041I0095 wo3 zhi1 dao4 ta1 bu4 xiang3 gen1 wo3 you3 gua1 ge2
+20170001P00298A0007 tian1 tian1 bao4 zhe5 xiao3 mian2 yang2 mei2 you3 hao3 xiong1 la5
+20170001P00176I0119 wo3 ri4 shang4 ci4 zi1 xun2 guo4 feng1 xiong1 de5
+20170001P00397I0042 zhi2 jie1 dang1 mo4 xiong1 qun2 chuan1
+20170001P00376A0046 fan4 shen1 qiong2 zai4 ren2 min2 yi1 yuan4 ren4 shi5 zhou1 ling2 er2
+20170001P00398A0031 ni3 qing1 yi4 xiang1 xin4 bie2 ren2 rang4 wo3 geng3 geng3 yu2 huai2
+20170001P00089A0027 lao3 ban3 niang2 shen1 huai2 liu4 jia3
+20170001P00391A0015 wei4 shen2 me5 jiu4 bu4 neng2 yu2 kuai4 de5 wan2 shua3 le5
+20170001P00243A0040 dai4 le5 zheng3 zheng3 yi2 ge4 xiao3 shi2 cai2 qi3 lai2
+20170001P00053I0119 wo3 jiu4 wen4 chen2 jun1 de5 ka3 zen3 me5 bu4 neng2 yong4 le5
+20170001P00193A0013 jiao3 tai4 hua2 lao3 shi4 shuai1 dao3 er4 shi2 fen1 zhong1 nv3 sheng1 zai4 tou2
+20170001P00114A0051 dan4 shi4 ye3 jiu4 yun3 xu3 fan2 yi1 hui4 er5
+20170001P00293I0015 xiao3 xue3 jie3 jie3 gei3 wo3 chang4 ge4 sheng1 ri4 kuai4 le4 bei5
+20170001P00286I0044 yue4 zhun3 bei4 lian4 zi4 he2 you2 yong3
+20170001P00191A0027 jia1 xiang1 hen3 duo1 ci2 ni3 men5 du2 chu1 lai2 bu4 yi2 ding4 zhi1 dao4
+20170001P00226I0084 ru2 guo3 pai2 chu2 zhe4 ge4 qian1 wan4 fu4 weng1 ban1
+20170001P00116I0081 wu3 xuan1 xian4 ren2 min2 yi1 yuan4 lai2 dian4 hua4
+20170001P00366A0017 mei2 you3 ren2 yun3 xu3 wo3 zai4 xiang4 hai2 zi5 yi1 yang4
+20170001P00106A0004 ba3 mao1 bo1 dao4 huai2 li3 lai2
+20170001P00397A0007 bu4 zhun3 zai4 tui1 ci2 la5 shuo1 hao3 le5
+20170001P00422A0030 ba3 ni3 dian4 ci2 bo1 xi2 ti2 ce4 jie4 wo3
+20170001P00423A0111 sheng4 dan4 jie2 kuai4 le4 zhu4 ni3 zao3 ri4 huai2 yun4
+20170001P00288A0019 ran2 hou4 cong2 wen1 zhou1 nan2 da1 che1 dao4 xia2 pu3
+20170001P00238A0096 na4 bi4 xu1 de5 ci2 xi1 zhe4 bian1 zhu4 su4 biao1 zhun3 shi4 duo1 shao3
+20170001P00330I0037 ni3 lia3 qin1 wen3 yong1 bao4 ma5
+20170001P00306I0068 mei2 you3 yao1 zhui1 guan3 xia2 zhai3 zheng4 xiang1 guan1 zheng4 zhuang4 zhe3
+20170001P00399A0071 hui4 tong1 guo4 ni3 de5 yun3 xu3 gen1 ni3 gou1 tong1
+20170001P00341A0061 kao3 yan4 yi3 hou4 xiao3 xun1 jiu4 shi4 fu4 qun2 zhu3 le5 o2 o2
+20170001P00309I0115 ta1 shuo1 deng3 chu1 yuan4 qu4 ta1 jia1 wan2 ba5
+20170001P00239A0086 sha2 shi2 huai2 shang3 qu4 he2 fei2 sha2 shi2 hui2 lai2
+20170001P00133A0106 hao3 huang2 mei3 tu2 xiu4 xiu4 ting3 fang1 bian4 de5 wa2
+20170001P00054A0106 qun2 deng3 ji2 da2 dao4 jun1 zhang3 jia4 ge2
+20170001P00117I0080 ye4 mo2 xia2 chi1 zhan4 lie4 kai1 wei3 qu1 hai2 shi4 jia1 li4 lai2 zhe5
+20170001P00233A0026 qian1 yue1 zhun3 bei4 he2 tong2 xu1 yao4 duo1 zhang3 shi2 jian1 a5
+20170001P00347I0113 wo3 hai2 ning4 yuan4 qu4 na3 er2 hai2 an1 yi4 dian3 ne5
+20170001P00074I0028 ni3 yao4 shi4 que1 ai4 huo4 zhe3 ye3 shi4 yi1 yang4 wan2 jia1
+20170001P00390A0031 er2 qie3 shi4 mo4 li4 xue3 qin1 qi1 kai1 de5
+20170001P00390A0050 jin1 nian2 yao4 hui2 lai2 shua3 mei4 mei4 wo3 shi4 ni3 ge1 ge1
+20170001P00224A0089 ni3 you4 bu2 shi4 he2 deng4 kun1 yi1 yang4 guai3 le5 liang2 jia1 fu4 nv3
+20170001P00012I0097 su4 she4 yi4 qi3 de5 ri4 zi5 wo3 si4 hu1 zong3 shi4 zai4 que1 xi2
+20170001P00177I0065 jin1 ri4 you4 qu4 ban4 guai1 nv3 a5
+20170001P00241A0092 mei4 mei4 ni3 na4 li3 hai2 you3 shen2 me5 jun1 qun2 ma5
+20170001P00290I0027 wo3 nai3 nai3 cong2 chuang2 shang4 xia4 lai2 le5
+20170001P00300I0094 ta1 men5 gei3 guai1 zi5 guo4 de5 shi4 nong2 li4 de5 sheng1 ri4
+20170001P00075A0117 wo3 zuo4 ni3 de5 di4 si4 zhe3
+20170001P00010I0080 shen1 ti3 you3 xie1 pi2 juan4 yu3 shuai1 lu4
+20170001P00108A0069 shou3 dong4 jia1 dian4 xiao4 ci2 kang1 fu4 zhan4 li4 chuang2
+20170001P00185A0059 na4 zhi3 chun3 huo4 ya2 chi3 hai2 bu4 zen3 me5 lou4
+20170001P00404A0075 peng2 ying1 zi5 zuo4 shen2 me5 shi4 hai2 yao4 shei2 yun3 xu3 ma5
+20170001P00120I0040 lao2 shan1 jun1 gong1 zhan4 dai4 lai2 de5 yi1 lun2 hong2 ri4 zhao4 bi4 hai3
+20170001P00330I0068 bei3 jing1 shi4 di4 zhi4 gong1 cheng2 kan1 cha2 yuan4 zhang1 yong3
+20170001P00311A0046 na4 jin1 tian1 wan3 shang4 wo3 men5 xi3 yuan1 yang1 yu4
+20170001P00021I0012 zhe4 shi4 wo3 ba4 de5 peng2 you3 lin2 ju1 jia1 de5 che1
+20170001P00177A0099 na3 tian1 wo3 pei2 ni3 he1 er4 bao3 de5 ya2 hai2 teng2 zhe5 ne5
+20170001P00138I0002 wan4 yuan2 yi3 shang4 zhe3 yi2 ding4 de5 sheng1 huo2 bian4 li4 he2 bu3 zhu4
+20170001P00432A0103 wo3 ba1 dian3 jiu4 shang4 ban1 er2 le5 huang2 qing1 ye3 shi4 shan1 li3 de5 wa2
+20170001P00101A0048 ke3 shi2 jian1 bu4 yun3 xu3 wo3 men5 liao2 liao2 tian1
+20170001P00344A0065 wo3 sun1 xue3 gan3 jin3 de5 jia1 wo3 kuai4 dian3
+20170001P00442A0019 weng4 an1 dao4 fu2 quan2 zui4 hou4 yi1 ci4 che1 ji3 dian3
+20170001P00148I0047 weng1 er3 sen1 zai4 na3 er2 ne5
+20170001P00325A0055 wei4 shen2 me5 bu4 zhun3 ni3 wan2 er2 shou3 ji1 ne5
+20170001P00172I0058 na4 ka3 qi2 se4 huo4 zhe3 jun1 lv4 se4 ne5
+20170001P00026I0116 ming2 tian1 yao4 jun1 xun4 le5 wo3 hai2 mei2 zhun3 bei4 hao3
+20170001P00334I0031 zhe4 liang3 tian1 qu4 xiang1 gang3 le5
+20170001P00343I0058 huang2 de2 xuan1 shi4 bu2 shi4 neng2 cha2 fen1 le5
+20170001P00063I0051 da4 xue3 guo4 hou4 qin2 huang2 dao3 da4 er4 de5 shi2 hou4
+20170001P00080A0109 zui4 mang2 de5 shi4 jiu4 shi4 he2 lao3 po2 yue1 hui4
+20170001P00166I0023 zi4 ji3 ci2 zhi2 bu4 gan4 cong2
+20170001P00200A0112 ku3 ming4 de5 yuan1 yang1 shen2 me5 shi2 hou4 cai2 neng2 ao2 chu1 tou2 ya5
+20170001P00258I0099 shen2 feng4 huang2 xiong1 di4 ni3 hao3
+20170001P00424A0041 wo3 ming2 tian1 zhun3 bei4 qu4 san1 men2 xia2 de5
+20170001P00211A0085 na4 ni3 hai2 bu4 ru2 qi1 wang4 li3 zong1 rui4 shi4 bei4 yuan1 wang3 de5
+20170001P00364A0094 ta1 men5 dou1 bu4 yun3 xu3 wo3
+20170001P00091I0082 xin4 nv3 jun1 sheng1 ri4 shi4 duo1 shao3 hao4 a5
+20170001P00277A0100 na4 tian1 wo3 zai4 an4 xiang1 dou1 yu4 dao4 yao2 jia1 le5
+20170001P00259I0080 qian1 yue1 ri4 qi1 nian2 yue4 ri4
+20170001P00189I0018 wo3 yao4 zhao3 ge4 zi4 cheng1 jie3 huo4 zhe3 ye2 de5
+20170001P00204I0068 qing1 jun1 you3 hen3 duo1 wu3 qi4 ri4 jun1 jian4 dou1 mei2 jian4 guo4
+20170001P00198A0036 kai1 ye4 you3 zeng4 song4 guang1 lin2 shi4 ni3 huai2 yun4 de5 shi2 qi1 ma5
+20170001P00405I0114 wan3 jiu3 dian3 di3 da2 guang3 zhou1 bai2 yun2 ji1
+20170001P00135A0052 lin2 jun4 yong3 yun2 cheng2 qu1 dou1 yang2 zhen4 wa3 tang2 cun1
+20170001P00364I0035 qing3 gai3 hao3 tian1 qing2 jun1 tuan2 de5 ni4 cheng1 zai4 shang4 mai4
+20170001P00133I0062 zhang1 li4 juan1 shuo1 wang2 chen2 de5 qun2 zi5 duan3
+20170001P00103A0109 zhuan4 dao4 qian2 ni3 sheng1 ri4 song4 ni3 ge4 chong1 qi4 wa2 wa2
+20170001P00018I0033 ai1 ya1 ben3 lai2 jiu4 dui4 wo3 xiong1 guo4
+20170001P00304I0092 na4 zhi3 neng2 shuo1 qun2 zhong4 de5 yan3 jing1 shi4 xue3 liang4 de5 le5
+20170001P00325A0009 hai2 shi4 zhun3 bei4 ru4 dang3 xuan1 shi4
+20170001P00358I0079 guai3 da4 shi1 bu4 zen3 bao4 yuan4 gong1 zuo4 a5
+20170001P00379A0024 qin1 ai4 de5 xue3 yin2 bao3 bei4 lao3 po2 lao3 gong1 de5 yan3 zai4 wang4
+20170001P00265A0048 ba3 wo3 men5 er4 qu1 gu1 niang2 dou1 guai3 qi3 pao4 le5
+20170001P00433A0078 xia1 hu2 che3 guang1 gei3 zhe4 you4 xiang3 ai1 xiong2 lei4
+20170001P00245A0019 ge2 a1 yi2 jiu4 ci2 gong1 zou3 le5
+20170001P00201A0118 lao3 niu2 mang2 quan2 min2 ying1 xiong2 feng1 hao4 le5 ni3 bie2 yong4 gua4 le5
+20170001P00248I0055 shi4 qu1 xun2 you2 pin3 pai2 xuan1 chuan2 quan2 jing3 tian1 chuang1
+20170001P00243I0046 xuan1 zhun3 bei4 he2 ta1 ma1 hui2 huai2 hua4
+20170001P00002I0035 dao4 di3 shi4 ji3 hao4 li3 qing1 lin2 huang2 xue3 wen2
+20170001P00367I0075 wo3 zhen1 xin1 bu4 xiang3 re3 ni3
+20170001P00019A0067 shu1 shu1 wo3 men5 bu4 yue1 ju2 zi5 dou1 gei3 ni3
+20170001P00359A0115 ao4 lin2 gu4 ke4 ding1 xue3 hong2 lin2 ba1 ke4 cheng2
+20170001P00326A0098 liu4 shi2 san1 jun1 da4 yuan4 xiang4 mu4 nei4
+20170001P00326I0038 wo3 hu2 wen1 gei3 de5 shi4 yang2 xuan1 chu1 wen3
+20170001P00085I0111 ken3 ding4 mei3 nian2 yi1 ci4 xun2 hui2 de5
+20170001P00201A0008 lao3 liang2 mei2 jiao4 gao1 san1 wo3 he2 xue3 xue3 jin1 tian1 qu4 mai3 piao4
+20170001P00133I0068 gao1 er3 fu1 qiu2
+20170001P00161A0021 jiu4 wei4 le5 ni3 de5 yi2 ge4 wei1 xiao4 huo4 zhe3 yong1 bao4
+20170001P00028A0087 jiu4 yue1 shang4 yun2 jie3 yi4 qi3 bei3 jing1 wan2 liang3 tian1
+20170001P00414A0074 mo4 gen1 ge1 ge1 che3 xie1 wai1 ba1 zi5
+20170001P00204A0108 liu2 yi4 ni3 you2 xiang1 ji4 de5 wo3 men5 de5 yue1 ding4 zao3 dian3 shui4
+20170001P00342A0020 qing3 yun3 xu3 wo3 zhe4 me5 jiao4 ni3 lai2 ke3
+20170001P00378A0119 shi1 xiong1 ni3 mei2 kan4 yu2 le4 bao4 dao4 ya5
+20170001P00356A0061 mang2 guo3 xiong1 mei4 men2 kou3 jian4 shi4 zai4 na4 fu4 jin4 ba5
+20170001P00082A0043 zhun3 bei4 si3 zai4 yong3 shi4 le5
+20170001P00215A0061 ni3 bi3 ren4 he2 shi2 hou4 de5 ni3 dou1 yao4 dan1 chun2 er2 yong3 gan3
+20170001P00302A0082 lao3 po2 ni3 shuo1 ke3 yi3 mo1 xiong1 de5 xia4 mian4
+20170001P00065I0111 yao4 bu4 yi1 lu4 ben3 lai2 yue1 le5 huang2 ping2 shi2 jian1 bu2 dui4
+20170001P00025I0088 er2 qie3 wo3 mai3 de5 yong3 yi1 hao3 chou3
+20170001P00373A0105 da4 jiang4 jun1 qu4 gei3 can1 ting1 gei3 zheng4 zhan4 ge4 zuo4
+20170001P00298I0063 bu2 shi4 wu3 xia2 zi1 shen1 ai4 hao4 zhe3
+20170001P00256I0065 dao4 shi2 hou4 kan4 kan4 neng2 bu4 neng2 ti2 qian2 yu4 yue1 chuang2 wei4 ba5
+20170001P00089I0026 ke3 bu4 ke3 yi3 qing3 wo3
+20170001P00162I0109 biao1 di4 che1 you4 qian2 lun2 tai1 ji2 gang1 quan1 shou4 sun3
+20170001P00119A0007 tou2 zhua1 zhu4 bei4 zi5 gun3 yi1 quan1
+20170001P00178I0031 xun2 qiu2 yi2 ge4 wen1 nuan3 de5 huai2 bao4
+20170001P00082I0057 ta1 hai2 wen4 qu4 qin2 huang2 dao3 ren2 jun1 san1 bai3 kuai4
+20170001P00122I0054 wo3 zhun3 bei4 zhang1 xing2 jun1 chuang2 guo4 ni3
+20170001P00425A0053 qun2 li3 mei3 nv3 yue1 hui4 gan4 ma5 ba3 wo3 dai4 shang4
+20170001P00019A0075 bao4 xue3 dui4 feng1 bao4 ying1 xiong2 jiu4 shi4 chun2 yu2 le4 xiang4 you2 xi4
+20170001P00245I0106 shui4 wan2 le5 mei2 zhun3 qu4 you2 yong3 ne5
+20170001P00351I0069 wo3 hai2 te4 yi4 jian4 le5 yi2 ge4 han4 bing1 you2 yong3 qun2 ne5
+20170001P00110I0051 nv3 xing4 huan4 zhe3 zai4 zhi4 liao2 zhong1 huai2 yun4 zhe3
+20170001P00196A0099 jiu4 qian2 ji3 tian1 wo3 he2 liu3 mei4 mo2 er3 you2 yong3 ni3 mei2 qu4
+20170001P00128A0062 huang2 qun2 xia2 na4 ni3 dang1 chu1 wei4 shen2 me5 yao4 gen1 wo3
+20170001P00218A0056 ni3 hua1 qian2 shou3 da4 bu4 da4 neng2 bu4 neng2 guo4 guan4 qiong2 ri4 zi5
+20170001P00406A0085 zhe4 ge4 hui2 da2 hen3 jiao3 hua2 ni3 mei2 you3 er4 xuan3 yi1
+20170001P00170A0002 wo3 zai4 hua2 dong1 tao2 ci2 pin3 shi4 chang3 xin1 shi2 dai4 guang3 chang3
+20170001P00139I0089 wo3 zuo4 chuang2 shang4 wan2 er2 you2 xi4
+20170001P00290A0038 lin2 jin1 zen3 me5 hai2 bu4 xia4 xue3 a5
+20170001P00230A0097 zhang3 da4 hui4 shi4 da4 ling3 gao1 liang2 sui4 wei3 ba1 de5 gong1 zhu3 mao1
+20170001P00300I0046 ye3 bu4 zhi1 dao4 kong1 jian1 li3 fa1 yuan4 de5 shi4 na3 zhi3 xiong2
+20170001P00157I0015 ni3 men5 zai4 qun2 li3 liao2 na4 qian2 mian4 xiong1
+20170001P00030I0098 guai4 bu4 de2 huai2 nian4 huang2 gong1
+20170001P00152A0107 mian4 shang4 yun3 xu3 xiao3 po4 chan3
+20170001P00049I0099 wo3 pang2 bian1 shi4 weng1 hong2 da2
+20170001P00287I0061 jue2 bu4 yun3 xu3 nan2 zi5 chu4 peng4
+20170001P00284A0003 ru2 guo3 ni3 hai2 zai4 wei4 zhe4 shi4 geng3 geng3 yu2 huai2
+20170001P00359A0015 qu4 zheng4 fu3 ba5 huo4 zhe3 ya2 ke1 yi1 yuan4
+20170001P00269A0045 ni3 jiu4 gan3 jin3 gei3 wo3 ci2 zhi2 zheng3 ge4 xi2 fu4
+20170001P00385A0066 yun3 xu3 ni3 chou4 mei3 da3 li3
+20170001P00211I0064 bie2 ren2 liu2 xing2 de5 ci2 hui4 huo4 zhe3 yin3 hui4 de5 ci2 yu3
+20170001P00127I0004 wo3 yi4 ban1 he1 xiang1 yu4 huo4 zhe3 ning2 meng2
+20170001P00271A0043 di4 er4 tian1 nv3 ren2 dou1 ai4 shua3 xiao3 pi2 qi4
+20170001P00220I0085 na4 jiu4 deng3 wo3 gai4 zhu1 quan1 de5 shi2 hou4 zai4 zhao3 ni3
+20170001P00215A0015 ni3 bu4 gan3 yun3 xu3 zi4 ji3 zuo4 zhe5 zi4 ji3 xi3 huan1 de5 shi4
+20170001P00093I0119 wo3 jie4 gei3 wo3 xiong1 di4 de5 ka3
+20170001P00424A0092 li3 na4 he2 sha1 la1 bo1 wa2
+20170001P00214I0118 gong1 zi1 ye3 shuo1 gei3 wo3 shen1 qing3 de5 shi4 an1 yong3 zi1 xun2
+20170001P00178A0079 bu2 yao4 chong1 qi4 wa2 wa2
+20170001P00082A0046 wo3 fa1 xian4 zhe4 wei4 lao3 xiong1 shi4 bu2 shi4 gai1 qu4 yi1 yuan4 a5
+20170001P00369A0062 wang3 xiong1 di4 qing2 qu4 yin3 dao3
+20170001P00016I0085 ni3 peng2 you3 quan1 zen3 me5 shen2 me5 dou1 mei2 you3
+20170001P00062I0043 qi2 shi2 wo3 ting1 ri4 zhong1 wu3 yue1 zu3 ren2 qu4 chi1 fan4 ge2 bo1
+20170001P00180A0005 ren2 min2 ying1 xiong2 ji4 nian4 bei1 zhong1 nan2 hai3 qu4 le5 ma5
+20170001P00228A0083 sha2 xiong2 mao1 yan3 de5 de5 qun2 li3 you3 lao3 shi1
+20170001P00058A0066 ni3 men5 dian4 na4 kuan3 wa2 wa2 qun2
+20170001P00282A0112 ran2 hou4 you4 ceng4 le5 er4 shi1 xiong1 yi1 dun4 kuai4 can1
+20170001P00402I0023 yang2 lang2 da4 jue2 zhan4 zhi1 da4 fu4 weng1
+20170001P00175A0051 shi4 jie4 shang4 de5 fu4 weng1 wang3 wang3 dou1 zou3 le5 na4 tiao2 shan1 lu4
+20170001P00383A0006 hai2 shi4 ci2 xi1 ren2 hao3 zhe4 shi4 qi3 bu4 jia4 gun3 yuan3 ba5
+20170001P00280A0029 qu4 wo3 fang2 jian1 de5 na4 ge4 guo4 dao4 er2 guai3 jiao3 chu4
+20170001P00422A0010 ta1 ti2 le5 nai4 he2 nan2 dao4 bu4 yun3 xu3 wo3
+20170001P00433A0011 er2 qie3 guo2 nei4 de5 jun1 xun4 shi2 zai4 tai4 wu2 liao2
+20170001P00064A0058 wei4 le5 you2 yong3 de5 ri4 zi5
+20170001P00392A0109 ni3 hai2 xiang3 you3 di4 er4 ge4 huo4 xuan1 yu3
+20170001P00193I0008 wo3 qi1 dai4 zhe5 zhe4 yang4 de5 bo1 li5 qun2
+20170001P00177A0117 tuan2 zhang3 hai2 neng2 yun3 xu3 wo3 qu4 wu3 fang2 wan2 wan2 ma5
+20170001P00004I0078 dong4 jing3 shang1 dong1 jing1 shang4 ye3 ri4 ben3 xiong2 mao1
+20170001P00399A0111 dang1 ran2 bu4 neng2 mo1 jiu4 bu4 zai4 chuang2 shang4 pao3
+20170001P00075I0007 wo3 tui4 qun2 xiao3 xiong2 qi4 xiao1 le5 zai4 jin4 lai2
+20170001P00355I0002 ni3 tou2 ding3 gai4 shang4 jiu4 xing2 cheng2 yi2 ge4 guang1 quan1 ba5
+20170001P00012A0004 xi1 ban1 ya2 ye3 tao2 bu4 guo4 ni3 de5 mo2 zhao3
+20170001P00164I0034 tie1 ba5 hen3 duo1 ren2 shuo1 da4 xia2 pei4 yin1 bi3 huang2 da4 hao3
+20170001P00333I0049 zhun3 bei4 le5 er2 wo4 de5 yi1 fu2
+20170001P00403I0081 gai1 huai2 yi2 xian1 dao3 fa2 bo1 wen2 guan3 sun3 huai4
+20170001P00086I0047 wo3 men5 bu4 dui4 que1 re3 a5
+20170001P00076I0106 bo2 ai4 yi1 ri4 juan1 hua1 ming2
+20170001P00368I0043 ma3 shang4 ba3 qing2 kuang4 gen1 lao3 weng1 ta1 men5 fan3 ying4 yi1 xia4
+20170001P00385I0040 qing3 yun3 xu3 wo3 ruo4 ruo4 de5 shu4 yi1 xia4 zhong1 zhi3
+20170001P00119A0109 liu2 jun1 ni3 dao4 cheng2 du1 zai4 xun2 wen4
+20170001P00062A0101 feng4 jie3 ming2 zhi1 dao4 wo3 bu4 zen3 me5 shua3 qun2 de5
+20170001P00226I0010 chi2 dao4 zhe3 bu4 yun3 xu3 shang4 gong1
+20170001P00353A0011 wai4 mian4 mai3 de5 shi2 hou4 you3 shi2 hou4 shua1 ka3 ci2 xing4 shua1 bu4 shang3 qu4
+20170001P00198A0007 kai1 che1 de5 yong3 yuan3 bu4 dong3 shou3 tao4 dui4 wo3 men5 qi2 che1 ren2
+20170001P00284A0034 ru2 guo3 ni3 chi1 xiang4 wu3 che1 jin4 ba1 qi1 jiang1 jun1
+20170001P00080A0039 wo3 gang1 cai2 you1 ya3 de5 yi1 shuai3 tou2
+20170001P00120A0004 beng4 beng4 da2 da2 de5 jiu4 yuan4 yi4 chu1 qu4 shua3
+20170001P00381A0079 na4 hai2 zai4 xia4 xue3 mei2 ma3 yun2 yun2
+20170001P00396I0100 ni3 zhun3 bei4 bao4 shen2 me5 shi2 hou4 de5 you2 yong3 yue4 ka3 a5
+20170001P00272A0095 wo3 mei2 you3 jun1 xun4 fu2 a5 yu2 le4 yu2 le4
+20170001P00126A0025 wan2 you2 xi4 ka3 ye3 zhi3 you3 bu4 fen4 you2 xi4 you3 you1 hua4
+20170001P00091I0096 gun3 dong4 zhou2 cheng2 run4 hua2
+20170001P00190I0085 jiu4 shi4 hui4 juan1 na4 ge4 yi1 yuan4
+20170001P00314I0016 shao3 liang4 nian2 tu3 kuang4 wu4 qie3 fen1 bu4 ji2 bu4 jun1 yun2
+20170001P00394A0087 shi4 a5 yin1 wei4 ta1 bu4 jing1 guo4 wo3 yun3 xu3 jiu4 yong4 wo3
+20170001P00103I0043 ru2 guo3 you3 huang2 xiao3 xiong2 pao3 dan1
+20170001P00279A0078 ni3 yao4 shi4 hui2 qu4 le5 wo3 jiu4 qu4 jun1 lin2 le5
+20170001P00217I0011 wo3 mai3 le5 yi1 tiao2 fen3 se4 bai2 bo1 dian3 de5 zhang3 qun2
+20170001P00188A0024 hu2 nan2 ken3 ding4 hui4 xia4 xue3 la5 zhong1 yu2 hui2 lai2 shen1 zhen4 le5
+20170001P00073A0104 yang2 guang1 yi1 zu2 huang2 po1 qu1
+20170001P00372I0108 bian4 si4 chu4 xun2 zhao3 ta1 xiang3 yao4 de5 chong1 lang4 hua2 ban3
+20170001P00230A0039 gao1 pin2 ci2 a5 qi2 ta1 lan3 de5 bei4 le5
+20170001P00177A0100 ni3 hao3 mei2 shang4 ban1 ma5 ni3 lao3 jia1 na3 de5
+20170001P00395A0061 shuo1 chu1 lai2 ye3 hui4 shi4 huai2 dian3 er2 yi1 dui1 wang3 yi4 dong1 xi1 ne5
+20170001P00415A0079 mo4 xie3 qian4 le5 yi1 da4 dui1 gai4 ci2 bi3
+20170001P00074A0097 di4 san1 zhe3 zai4 na4 ge4 shang4 mian4 tang3 zhe5 ne5
+20170001P00235I0089 ta1 yao4 wo3 qu4 huang2 shan1 na4 li3 pao3 yi1 quan1 shi4 shi4 kan4
+20170001P00132A0079 chao1 ji2 ying1 xiong2 lian2 meng2 fu4 chou2 zhe3 di4 er4 ji4
+20170001P00038I0035 lu4 wa2 ni3 zhun3 bei4 chong1 duo1 shao3
+20170001P00393I0050 zui4 jin4 zhun3 bei4 xue2 xi2 you2 yong3 qu4
+20170001P00225A0086 nin2 bu2 shi4 shuo1 wo3 jie3 jie3 shi4 nin2 zui4 hao3 de5 xiong1 di4 me5
+20170001P00144A0106 bu4 neng2 you3 bu4 wen2 ya3 de5 ci2 yu3
+20170001P00150A0112 wo3 zhi3 hui4 shun4 zhe5 bie2 ren2 de5 jie2 zou4 an1 pai2
+20170001P00159I0051 huang2 shou3 juan1 ma3 shang4 hui2 wu3 han4 fu4 xi2 kao3 yan2
+20170001P00419A0063 ba3 wo3 dang1 xiong1 di4 jiu4 qu4 kong1 jian1 yi1 er4 tiao2 dou1 zhuan3 fa1
+20170001P00303I0106 wo3 zhi1 dao4 wo3 you4 re3 ni3 bu4 kai1 xin1 le5
+20170001P00195A0022 jin4 da4 men2 you4 guai3 dao4 shu3 di4 er4 ge4 lou2 shi4 jiu3 hao4 lou2
+20170001P00304A0011 li3 zheng4 xin4 shi1 xiong1 wo3 fa1 xian4 nin2 tai4 du4
+20170001P00238I0113 wo3 zhi1 dao4 ni3 jiu4 shi4 xia1 che3 dan4 de5
+20170001P00395I0007 dou1 shi4 yi1 qun2 ke3 lian2 de5 wa2 a5
+20170001P00333A0039 mo4 che3 shang4 jia1 ge1 ma5 jiu4 shi4 ni3 zi4 ji3 xiang3 he2
+20170001P00062I0018 jiu4 wen4 yi1 xia4 huang2 qiao3 jun1 zen3 me5 zuo4 che1 qu4
+20170001P00098I0083 a1 huang2 he2 dan1 jie3 xia4 qu4 shuo1 wo3 you3 yue1 yao4 dai4 shang4 ta1 men5
+20170001P00239A0077 zhun3 bei4 ge4 bing1 gui4 ba5 you3 guo4 qi1 de5
+20170001P00393I0086 na4 ge4 xiang4 shi4 qun2 xuan1 chuan2
+20170001P00167I0008 ri4 jun1 liu2 lan3 liang4 you3 duo1 shao3
+20170001P00107A0042 ni3 yue4 jing1 bu4 zhun3 hai2 zhe4 me5 rong2 yi4 huai2 yun4
+20170001P00159A0081 zhun3 bei4 chu1 xin1 ying1 xiong2 le5
+20170001P00006I0064 guo2 xiao3 jin1 ri4 you4 zou3 qu4 ci2 gong1
+20170001P00229I0021 ta1 de5 xin1 li3 mian4 yi4 zhi2 dou1 hai2 you3 huang2 xue3
+20170001P00135I0052 weng4 an1 di4 tu2 shang4 ni3 zhao3 bu2 dao4 de5
+20170001P00160I0034 shou3 jia1 xin1 sheng1 er2 you2 yong3 fu3 chu4 shi4 zai4 yi1 yuan4 luo4 cheng2
+20170001P00408I0053 na4 ni3 qu4 yao2 gun3 de5 hua4 qi3 bu4 shi4 geng4 yao4 que1 yang3 la5
+20170001P00391I0114 shei2 you3 quan2 jun1 tong3 chou2 de5 yi1 yuan4 ming2 dan1
+20170001P00024I0056 bu4 zao4 ren2 me5 ge1 ci2 bu2 shi4 li3 ke4 qin2 de5 ma5 o2
+20170001P00277A0029 na4 wa2 ha1 ha1 chou4 wen2 zi5 dou1 bei4 xun1 yun1 le5
+20170001P00356I0020 ni3 ye3 lai2 wan2 xuan4 wu3 san1
+20170001P00021I0073 huan1 ying2 qiu1 yi4 nong2 xue3 er2 jin4 qun2
+20170001P00076A0027 dou1 dao4 ma3 tou2 le5 huang2 hua2 gang3
+20170001P00148I0091 wo3 zhe4 li3 you3 jun1 xun4 de5 zhao4 pian1 lao3 po2 kan4 bu4
+20170001P00153A0033 e2 luo2 si1 nv3 shi4 shou3 biao3 e2 luo2 si1 jun1 yong4 miao2 zhun3 jing4
+20170001P00163I0008 huang2 jin1 xian4 huo4 de5 he2 yue1 dan1 wei4
+20170001P00354A0014 mei3 nv3 gen1 ma1 ma1 dou1 shi4 yun2 nan2
+20170001P00146I0081 wo3 zai4 wo3 men5 yuan4 zi5 li3 kan4 le5 yi1 quan1 bei4 bei4 zui4 hao3
+20170001P00434A0071 fan3 zheng4 zui4 jin4 gei3 ta1 jiao1 liu2 you3 dai4 gou1
+20170001P00426A0080 lian4 che1 na4 li3 shua3 shun4 bian4 jiao4 xue2 yuan4
+20170001P00366A0058 gao1 te4 bu4 yun3 xu3 zhan4 ren2
+20170001P00181A0103 ha1 ha1 xue1 hai2 zai4 wai4 mian4 ne5 he2 tong2 shi4 he1 dian3 er2 xiao3 jiu3
+20170001P00242I0095 tian1 jin1 da4 xue2 jun1 shi4 jiao1 tong1 xue2 yuan4
+20170001P00107I0094 wo3 men5 gong1 si1 de5 tong2 shi4 zhun3 bei4 sheng1 di4 er4 tai1 le5
+20170001P00299A0032 wo3 ban1 de5 xiong2 er4 shai4 ta1 nan2 peng2 you3 de5 zhao4 pian1
+20170001P00282A0088 rang4 ni3 chi1 xue3 dao4 jia1 le5 jin1 tian1 fei1 shen3 yang2
+20170001P00304A0008 li4 jiang1 yue1 pao4 zhi1 lv3 ti2 shang4 xing2 cheng2 kuai4
+20170001P00289A0084 ni3 ji4 bu2 ji4 de2 wang2 cun1 xue3 gei3 ni3 de5
+20170001P00139I0009 ni3 tang2 er2 huang2 zhi1 de5 dai4 shang4 ni3 de5 dong1 xi1 zou3
+20170001P00289A0053 wo3 ye3 bu4 zhi1 dao4 tong1 zhi1 xia4 wang2 lin2 xuan1 ba5
+20170001P00141I0079 tuan2 jie2 yi1 zhi4 ban4 hao3 qun2 zhong4 yong1 hu4 de5 shi2 shi4 hao3 shi4
+20170001P00138I0033 ya4 ka3 ba1 quan2 jian1 di2 jun1 yi1 ge4 ban4 man3 bian1
+20170001P00318A0066 ni3 qu4 quan1 yuan2 cun1 zi4 ji3 zhu4 yi4 an1 quan2
+20170001P00263I0088 ri4 jun1 you3 xiao4 bao4 an4 liang4
+20170001P00226I0083 wo3 ai4 ni3 wen3 ni3 hao3 xiang3 ni3 juan1 er2
+20170001P00094I0101 jin1 tian1 wo3 gan4 xiong1 di4 guo4 sheng1 ri4
+20170001P00039A0108 wo3 zhe4 ge4 fen1 shu4 cai2 che3 le5 ne5
+20170001P00103I0045 gu1 ji4 shi4 qian1 wan4 fu4 weng1 a5
+20170001P00379I0055 hou2 zi5 yao4 qu4 zhong1 xin1 guang3 chang3 wan2 er2
+20170001P00412A0037 yin1 wei4 nv3 sheng1 xiao3 shi2 hou4 wan2 yang2 wa2 wa2 bi3 jiao4 duo1
+20170001P00143I0092 wo3 mei2 bang1 xiao3 weng1 jiao4 o2
+20170001P00252A0113 gei3 tui1 jian4 xia4 wo3 zai4 nan2 fang1 he2 nan2 qiong2 xiang1 pi4 ye3
+20170001P00319I0114 cong2 chuang2 shang4 pu4 shuai4 shuai1 shuai1 shi4 a5
+20170001P00425A0007 ni3 pei4 he2 ta1 jiu4 xing2
+20170001P00278I0038 jing4 zhong1 de5 a1 shan3 zhuo2 yi1 shen1 xue3 bai2 sha1 qun2
+20170001P00042I0086 dan4 xiao3 min3 min3 hua4 ru2 guo3 huang2 gui4 yong3 jiu4 xi4
+20170001P00028A0016 sheng1 yin1 ya3 de2 you3 ci2 xing4 hei1 de5
+20170001P00326A0051 wan2 quan2 shuai1 dong1 ri4 na4 ji3 tiao2 lian3 jiu4 xiang3 cai3 ji3 jiao3 dong1
+20170001P00101A0084 wo3 mei2 qian2 a5 qiong2 de5 zhun3 bei4 chi1 xi2 lai2 le5
+20170001P00241A0015 guang3 xin4 jun1 da2 lao3 zong3 ta1 shi4 ting3 re4 qing2 de5
+20170001P00023A0047 huo4 zhe3 shi4 gen1 ben3 bu4 na2 wo3 dang1 xiong1 di4 hao3 ma5
+20170001P00164A0084 wo3 huai2 yun4 hou4 xi1 gai4 qian2 hou4 yang3 yang3 de5 bu4 xing2 le5
+20170001P00360I0004 bu4 quan4 ta1 hao3 hao3 xue2 xi2 hai2 song3 yong3 ta1 wan2 er2
+20170001P00381I0096 pang4 zi5 shuo1 wo3 men5 jie2 hun1 ta1 jiu4 gen1 wo3 men5 xing4
+20170001P00310I0021 wo3 men5 qun2 li3 nei3 ge4 zai4 wan2 er2
+20170001P00184A0037 wan3 shang4 he2 li4 min3 ta1 men5 yue1 hui4
+20170001P00212A0074 na4 wo3 xian1 xie4 le5 a5 luo2 jian4 de2 li3 yun2 jun1 shang4 ban1 le5
+20170001P00362A0077 shi4 jing3 ti4 zhe4 ge4 ci2 shi4 wang3 luo4 ci2
+20170001P00157A0105 bu4 hao3 ba2 hai2 yao4 ba3 ya2 chuang2 qie4 kai1
+20170001P00166I0004 tu2 an1 weng1 de5 ren2 cai2 hui4 liu2
+20170001P00375A0042 qing1 bai2 hui3 diao4 jiao4 shui3 wa2 hui3 le5
+20170001P00002I0013 jing3 wei4 lian2 ban1 zhang3 ri4 chang2 jun1 dui4 gong1 zuo4 yi4 wu4 bing1
+20170001P00081I0053 tang3 zai4 chuang2 shang4 bei4 dan1 ci2
+20170001P00130I0075 wu3 yuan4 dui4 mian4 de5 you2 yong3 guan3
+20170001P00022I0037 rong2 qiao2 you2 yong3 chi2 nian2 ka3
+20170001P00045I0052 jian3 yue1 de5 tiao2 wen2 qun2 zi5
+20170001P00347I0117 zi1 bo2 ri4 yong4 ci2 chang3 jia1
+20170001P00177I0086 wo3 men5 de5 juan1 kuan3 pai2 qing3 zuo4 cheng2 hong2 di3 huang2 zi4 de5
+20170001P00067A0009 gei3 yu2 gang1 huan4 shui3 ne5 huai2 yun4 le5 bu4 gan4 huo2
+20170001P00101I0104 wo3 men5 gai1 nong4 ge4 er4 qun2
+20170001P00402A0053 wai4 po2 sheng1 ri4 jia1 li3 ren2 que1 shao3 le5 wo3 ta1 bu4 kai1 xin1
+20170001P00382I0056 piao4 zi5 yun3 xu3 di2 qing2 kuang4 xia4 xiang3 huan4 jiu4 huan4
+20170001P00197I0042 huai2 an1 feng4 huang2 guo2 lv3 cui1 zong3
+20170001P00151A0119 wo3 yi1 jiang3 hua4 ye3 weng1 weng1 weng1 de5 sheng1 yin1
+20170001P00176I0117 wo3 kao4 yao4 gai4 zhang1 hai2 de2 xie3 bao4 gao4 hai2 yao4 da2 bian4 a5
+20170001P00053I0061 wo3 ta4 ri4 qu1 de5 jun1 hun2
+20170001P00338A0023 wo3 de5 xin1 xiong1 mei2 na4 me5 kuan1 da4
+20170001P00095A0004 li4 shi3 ri4 zhi4 jun1 hui4 bei4 ya1 suo1 bao1 cun1
+20170001P00241A0114 bu4 zhun3 shua3 lai4 o2 ni3 zi4 ji3 shuo1 de5 bu2 hui4
+20170001P00144I0076 wo3 kan4 ren2 jia1 hen3 duo1 qun2 dou1 you3 gu1 gu1
+20170001P00036A0096 wo3 te4 huai2 yi2 ta1 shi4 huang2 yu3 cong1
+20170001P00018I0070 gao1 kao3 wan2 le5 zai4 man4 man4 qu4 shua3 jiang1 ba5
+20170001P00084A0103 shuo1 ming2 wo3 cuo4 ci2 zhun3 que4
+20170001P00385A0020 jiu4 chu1 xian4 ying1 xiong2 de5 ji4 cheng2 zhe3
+20170001P00399I0014 jia3 wu3 zhan4 zheng1 ri4 jun1 xu4 lie4
+20170001P00137A0047 yuan4 jia1 na4 mi3 she2 you2 jing1
+20170001P00393I0030 gun3 hui2 ni3 de5 zhao3 wa1 xing1 qiu2 qu4
+20170001P00103A0011 wo3 yi1 kai1 shi3 xiang3 huan4 huang2 gai4 de5
+20170001P00383A0030 yun3 xu3 wo3 zuo4 ge4 bei1 shang1 de5 biao3 qing2
+20170001P00083I0075 yang2 ya4 hui1 he2 chen2 rong2 jun1 de5 fen1 hong2 ka3
+20170001P00299I0017 zhu4 xiao1 de5 lai2 hua4 ri4 jun1 xia4 jiang4
+20170001P00277I0071 hua2 kuai4 yu3 shui3 ping2 qing1 zhi4 tan2 huang2 de5 yi1 duan1 xiang1 lian2
+20170001P00128I0006 nei4 meng2 gu3 chi4 feng1 shi4 weng1 niu2 te4 qi2 you2 zheng4 bian1 ma3
+20170001P00256I0009 ni3 tai4 qian1 xu1 le5 wo3 zai4 gen1 wo3 ge1 che3 gua3
+20170001P00166I0069 jin1 ri4 you4 qu4 yue1 hui4 a5
+20170001P00226A0013 neng2 bu4 neng2 gei3 wo3 yi2 ge4 wen1 nuan3 de5 yong1 bao4
+20170001P00080A0057 qu4 yun2 nan2 yu4 long2 xue3 shan1 de5 zhu4 yi4 shi4 xiang4
+20170001P00236I0039 ri4 fan4 jiu4 bu4 neng2 cheng2 quan1
+20170001P00068A0041 ma2 bi3 dou1 zao1 ri4 ge4 leng2 ge4 yong1 zhong3 ha1 ha1 ha1
+20170001P00202A0094 li3 yu3 xuan1 shi4 bu2 shi4 yu2 wen2 ke1 yu2 wen2 ke1 xi3 huan1 shei2
+20170001P00215A0070 ni3 cai2 shi4 xiao3 yin2 wa2 ne5 wo3 ma1 zhe4 ji3 tian1 qu4 he2 fei2
+20170001P00253I0107 huang2 jian4 yong3 jia1 you3 tian1 de5 dan1 zi5 mei2 gei3 wo3
+20170001P00024A0046 xian4 zai4 ta1 ye3 bu4 yuan4 yi4 wo3 ci2 zhi2 le5
+20170001P00299A0075 kuang2 feng1 lang4 die2 qiong2 zhui1 bu4 she3
+20170001P00295I0035 en1 dao4 yuan4 ti2 shi4 nin2 wang3 shang4 yu4 yue1 de5
+20170001P00374A0068 nao3 dai4 weng1 weng1 de5 shang4 cheng2 qiang2 xian1 bie2 dong4
+20170001P00412A0094 fan3 yi4 ci2 bu4 xiang3 bao1 yi4 ci2 na4 huang2 dao4 qiang2 di1
+20170001P00208I0019 xian4 zai4 lao3 po2 dou1 huai2 yi2
+20170001P00185A0002 wo3 jiu4 xiang3 zhi1 dao4 qi2 zhong1 you3 mei2 you3 ta1 er2 yi3
+20170001P00169A0095 gang1 cai2 da3 xue3 zhang4 qu4 le5 xian4 zai4 chi1 kou3 fan4 xie1 yi1 hui4 er5
+20170001P00271I0065 ni3 ren4 wei2 you3 ren2 hui4 yuan4 yi4 yong4 ma5
+20170001P00287A0102 lao3 xiong1 wo3 xian4 zai4 shang4 ni3 de5 wang3 pan2
+20170001P00126I0063 shei2 you3 quan2 jun1 tong3 chou2 de5 yi1 yuan4 ming2 dan1
+20170001P00218I0018 wo3 zhen1 de5 shi4 pa4 gei3 ni3 re3 ma2 fan2
+20170001P00048A0007 hao3 huai2 nian4 zhe4 xie1 ri4 zi5
+20170001P00390A0091 ni3 shi4 shi2 yang4 shuai1 a5 me5 ni3 du2 dao4 chu1 san1
+20170001P00129A0077 wo3 zhou1 mo4 feng1 kuang2 mi2 shang4 xuan1 yuan2 jian4 le5
+20170001P00115I0001 na4 ge4 xiao3 xiong2 bi3 ri4 ben3 er2 de5 shi4 tai4 gui4 le5
+20170001P00039I0103 ni3 you3 kong4 de5 hua4 qu4 bei3 yi1 huo4 zhe3 bei3 er4 wan2 er2 wan2 er2
+20170001P00278A0003 ni3 de5 xiang4 ji1 bi3 wo3 men5 de5 hao3 tai4 duo1 le5
+20170001P00092A0068 da4 xiao3 jun1 yun2 ren2 bu4 gao4 su4 ni3
+20170001P00160A0119 ni3 yi3 wei2 wo3 shi4 fu4 weng1 ma5 da4 mi3 zhe4 ge4 jiu4 hao3
+20170001P00158I0069 fu2 lan2 xue3 ya1 xiang1 yi2 ben3 cao3 shao2 yao4
+20170001P00076A0082 ge1 zhun3 bei4 hao3 dao4 guang3 da4 lun4 tan2 chang4 shuai1 ni3 men5
+20170001P00010I0106 ni3 wen4 xiong1 di4 wan2 xun2 xian1 ma5
+20170001P00093I0053 yi1 ming2 fa3 zhi4 xuan1 chuan2 zhi4 yuan4 zhe3
+20170001P00412A0082 ta1 men5 shen2 me5 shi2 hou4 qu3 xiao1 yun3 er2
+20170001P00410I0116 jiu4 shi4 ou3 er3 you2 yong3 he2 tiao4 wu3 yuan4 li3 qu4 nian2 fa1 de5
+20170001P00218A0068 ni3 jue2 de5 ni3 lao3 biao3 shi4 bai3 wan4 fu4 weng1 you3 yi4 yi4 ma5
+20170001P00074I0007 huang2 yong3 chao1 zhe4 li3 za3 mei2 ge4 zheng4 chang2 dian3 de5
+20170001P00026A0022 na4 dao4 ye3 shi4 en4 xiong2 shan1 sheng1 ri4
+20170001P00235I0052 jia3 fang1 wei4 an4 ben3 xie2 yi4 yue1 ding4
+20170001P00331I0098 wo3 zhe4 yi1 liang3 tian1 jiu4 zhun3 bei4 ci2 zhi2 le5
+20170001P00226I0033 xin1 bo2 quan1 gai1 gai3 zhong4 liang4
+20170001P00055A0081 huang2 shuang3 fa1 peng2 you3 quan1 er2 jiu4 fa1 wo3 yi2 ge4 ren2 kan4
+20170001P00233I0018 xiao3 xin1 ni3 lao3 po2 bei4 bie2 ren2 guai3 pao3 a5
+20170001P00127A0061 hai2 dai4 da4 yi1 xin1 sheng1 jun1 xun4 ne5 lao3 shuang3 le5
+20170001P00037I0086 ni3 zhun3 bei4 shi2 yi1 yue4 ci2 zhi2
+20170001P00442A0041 jiang1 yun2 shen1 he2 zi3 teng2 wan3 er3 yi2 xiao4
+20170001P00205I0118 hu4 jiang1 ri4 yu3 kai1 xin1 ci2 chang3 xia4 zai4
+20170001P00052I0059 bu2 yao4 he2 bai3 wan4 fu4 weng1 bi3
+20170001P00071A0100 ni3 wa2 wa2 zou3 shi1 diao4 zhe4 me5 jiu3
+20170001P00148A0113 huang2 se4 de5 xiong1 bi3 jiao4 da4
+20170001P00319I0049 lao3 nian2 ren2 chang2 wen3 er2 bu4 jian4
+20170001P00045I0047 qun2 zhong4 he2 jun1 dui4 di2 shi4 qi4 bu4 ke3 jian3
+20170001P00303A0068 li3 pei4 zhe4 suan1 shuang3 ni3 yao4 he2 wu2 shi4 xun1
+20170001P00126I0024 xun2 wen4 zhen1 dui4 xuan1 chuan2 fang1 mian4 you3 he2 ji4 hua4
+20170001P00254I0031 zui4 weng1 zhi1 yi4 bu4 zai4 jiu3 de5 yi4 de5 jie3 shi4
+20170001P00086A0021 kuai4 fang4 jia4 le5 ren3 ren3 ba5 zhi3 dai4 le5 xiong1 ka3 he2 bi3
+20170001P00038I0059 ren4 he2 ying1 xiong2 dou1 you3 hui1 huang2 de5 shi2 hou4
+20170001P00030A0104 wo3 hai2 shi4 guai1 guai1 kan4 wo3 de5 huang2 jin1 ba5
+20170001P00075A0026 ni3 bu4 pa4 huang2 he4 ban1 hei1 yan3 quan1
+20170001P00069A0118 na4 ni3 bang1 wo3 zai4 yue1 yi2 ge4 ba5 ai1 ni3 dao4 le5
+20170001P00272I0060 suo3 you3 zi1 liao4 jun1 wei4 zhuan3 yi2 dao4 min3 xing2 fa3 yuan4
+20170001P00035A0003 jian4 liang4 ya5 you4 bian1 de5 shi4 qun2 zhu3
+20170001P00175A0074 ni3 zong3 re3 ren2 jia1 nan2 wa2 er2 luan4 xiang3
+20170001P00226I0018 mei2 gei3 ta1 zuo4 xiong1 ka3 de5
+20170001P00393I0033 jin1 qi4 he2 ku4 zi5 shi4 qun2 li3 chu1 xian4 lv4 zui4 gao1 de5 ci2
+20170001P00335A0009 tang3 zai4 chuang2 shang4 guai1 guai1 de5 deng3 wo3 guo4 qu4
+20170001P00429A0061 zhi1 dao4 de5 hua4 hui4 yun3 xu3 ni3 hu2 zuo4 fei1 wei2 ma5
+20170001P00251I0101 wo3 ban4 de5 shi4 yi1 zhao4 wei3 de2 tong1 dian4 you2 yong3 ka3
+20170001P00214I0060 ben3 qun2 de5 mei2 shang4 gu1 gu1 o2
+20170001P00087A0107 shuai4 you4 dai1 meng2 ke3 neng2 shi4 wang4 jun1 de5 yang2 mi4
+20170001P00186A0089 na4 ge4 ci2 xiang2 de5 shuang1 yan3 pi2 kan4 chu1 lai2
+20170001P00112A0039 wo3 shi4 ge1 ci2 ai4 hao4 zhe3
+20170001P00386A0106 wo3 ji4 de5 shi4 lei4 si4 kai3 jia3 yong3 shi4 de5 ke3 yi3
+20170001P00191A0118 jia1 jin4 qu4 le5 ta1 shi4 ba1 dun4 jiang1 jun1 wo3 de5 shi4 bei4 bei4
+20170001P00089A0117 ni3 shi4 yao4 ya1 dao3 qun2 xiong2 me5
+20170001P00156A0001 lao3 weng1 you4 zai4 da3 wo3 de5 zhu3 yi4 le5
+20170001P00305A0056 liang3 xiong1 di4 a5 dou1 bu2 yao4 guo4 le5 sheng1 ri4
+20170001P00088I0021 huang2 ye4 shi4 ni3 men5 she4 ji4 de5 ma5
+20170001P00348I0095 kao3 lv4 dang1 qian2 huan4 zhe3 hu1 xi1 shuai1 jie2 ming2 xian3 de5 xian4 xiang4
+20170001P00330A0057 hao3 ji3 ci4 dou1 zhun3 bei4 ci2 zhi2 le5
+20170001P00423A0007 ge4 wei4 ya3 hu3 de5 ying1 xiong2 hao3 han4
+20170001P00186I0088 gong1 zuo4 ri4 wo3 bu4 jie1 shou4 yu4 yue1 wai4 chu1 wa5
+20170001P00092I0028 qun2 ni4 cheng1 de5 min3 gan3 ci2 hui4
+20170001P00126I0115 wo3 zai4 da4 xue3 ri4 zi5 jiu4 de5 na4 ge4 ren2
+20170001P00395I0035 bie2 ren2 zhun3 bei4 wan2 peng2 you3 le5
+20170001P00234A0043 ni3 da4 mang2 ren2 diao1 xiong1 wo3 de5 qun2 you3 ren2
+20170001P00347A0079 ni3 hai2 shi4 zhua1 jin3 ba3 bie2 ren2 shuai3 le5
+20170001P00135I0010 zhe4 qi2 shi2 shi4 jun1 dui4 yong4 de5 bo2 ji1 shu4
+20170001P00109A0117 wo3 ji4 de5 wo3 hai2 bei4 yong3 yang2 gao1 zhong1 lu4 qu3 guo4 ne5
+20170001P00263I0015 ni3 xian4 zai4 zhun3 bei4 huai2 yun4 le5 ma5
+20170001P00086I0110 ran2 hou4 ci4 ri4 zao3 chen2 dai4 zhe5 xiong2 mao1 yan3 er2 qu4 shang4 ban1
+20170001P00099A0007 nan2 dao4 shi4 wo3 shuo1 cuo4 le5 ma5
+20170001P00066A0067 bu4 zhun3 ren4 he2 ren2 jin4 ban1 xuan1 chuan2
+20170001P00025I0085 shuo1 shuo1 hao3 xiong1 sha2 mei2 gan4 ma5 ya5
+20170001P00371I0113 weng1 fan1 zhen1 de5 xi3 huan1 yang2 zhen4 ning2 ma5
+20170001P00330A0035 na4 ni3 xue2 xuan1 chuan2 hou4 qi1 hen3 zhan4 pian2 yi2
+20170001P00129I0109 wo3 shi4 er4 yuan4 xuan1 chuan2 ke1
+20170001P00253A0082 gou4 mai3 yi1 tao4 jin3 se4 yi1 zhuang1 gao1 ji2 qi2 xiong1 li3 fu2
+20170001P00063A0064 wo3 shi4 he2 yi1 qun2 diao3 si1 shang4 jun1 shi4 li3 lun4
+20170001P00362A0007 weng1 si1 ying2 shuo1 wo3 bu4 gen1 ta1 wan2
+20170001P00239I0022 ran2 hou4 zuo4 xiang3 bai3 wan4 fu4 weng1 zhi1 fu2
+20170001P00203A0031 lin2 zai3 jie4 ni3 dian3 qian2 ni3 lai2 xin1 dian4 yi1 xia4 bei5 xiong1 di4
+20170001P00107A0038 dian4 ci2 bo1 fang2 hu4 wei2 qun2 ye3 ting3 hao3 de5
+20170001P00077A0113 ke3 shi4 mei2 you3 na4 zhong3 yun3 xu3
+20170001P00307I0082 jiang1 yuan1 cha1 ru4 yu4 shi4 qiang2 shang4 de5 ka3 cao2
+20170001P00344A0098 dui4 bu4 qi3 qing3 yun3 xu3 wo3 xiao1 shi1
+20170001P00361A0015 kan4 kan4 you3 mei2 you3 jing4 pian4 mai4 wo3
+20170001P00120A0038 huo4 zhe3 jiao4 xiao3 xiong2 da4 shen2 bei4 fen4 ta1 de5 xi4 tong3 chu1 lai2
+20170001P00118A0068 tan4 guan4 dian4 ci2 fa2 you2 xiang1 gai4
+20170001P00283A0047 ren2 pin3 you3 na4 me5 cha4 ma5 nei3 xie1 xiong1 di4 a5
+20170001P00135I0066 na2 wo3 de5 pu1 gai4 he2 dian4 nao3 er2 yi3
+20170001P00126A0029 lian2 xu4 jun1 yun2 de5 weng1 weng1 sheng1
+20170001P00402I0035 sheng1 ri4 kuai4 le4 o1 huang2 di4 bi4 xia4
+20170001P00171I0037 ni3 zai4 xiang3 shen2 me5 a5 a5 a5 juan3 mao2 jun1
+20170001P00025A0096 kuai4 lai2 wan2 shua3 jin1 tian1 wo3 sheng1 ri4
+20170001P00019I0029 hai2 zai4 hu1 zhe4 xie1 le5 gai4 ci2 bi3
+20170001P00261I0051 ba3 yi3 qian2 zai4 tan2 huang2 chang3 de5 gong1 zuo4 ci2 le5
+20170001P00150I0069 jin1 ri4 qu4 qiong2 hua1 bi3 sai4
+20170001P00099I0099 wo3 zhun3 bei4 ming2 tian1 xia4 dao4 shuang1 jun1 de5

datalist/st-cmds/dev.wav.txt

@@ -0,0 +1,600 @@
+20170001P00142A0070 ST-CMDS-20170001_1-OS/20170001P00142A0070.wav
+20170001P00196A0072 ST-CMDS-20170001_1-OS/20170001P00196A0072.wav
+20170001P00337A0004 ST-CMDS-20170001_1-OS/20170001P00337A0004.wav
+20170001P00148I0088 ST-CMDS-20170001_1-OS/20170001P00148I0088.wav
+20170001P00131I0047 ST-CMDS-20170001_1-OS/20170001P00131I0047.wav
+20170001P00368I0013 ST-CMDS-20170001_1-OS/20170001P00368I0013.wav
+20170001P00367A0059 ST-CMDS-20170001_1-OS/20170001P00367A0059.wav
+20170001P00213A0115 ST-CMDS-20170001_1-OS/20170001P00213A0115.wav
+20170001P00035I0069 ST-CMDS-20170001_1-OS/20170001P00035I0069.wav
+20170001P00267I0078 ST-CMDS-20170001_1-OS/20170001P00267I0078.wav
+20170001P00344I0013 ST-CMDS-20170001_1-OS/20170001P00344I0013.wav
+20170001P00027I0050 ST-CMDS-20170001_1-OS/20170001P00027I0050.wav
+20170001P00090A0025 ST-CMDS-20170001_1-OS/20170001P00090A0025.wav
+20170001P00262A0029 ST-CMDS-20170001_1-OS/20170001P00262A0029.wav
+20170001P00316A0080 ST-CMDS-20170001_1-OS/20170001P00316A0080.wav
+20170001P00206A0047 ST-CMDS-20170001_1-OS/20170001P00206A0047.wav
+20170001P00350A0027 ST-CMDS-20170001_1-OS/20170001P00350A0027.wav
+20170001P00325I0013 ST-CMDS-20170001_1-OS/20170001P00325I0013.wav
+20170001P00071A0011 ST-CMDS-20170001_1-OS/20170001P00071A0011.wav
+20170001P00091I0112 ST-CMDS-20170001_1-OS/20170001P00091I0112.wav
+20170001P00091I0117 ST-CMDS-20170001_1-OS/20170001P00091I0117.wav
+20170001P00353I0057 ST-CMDS-20170001_1-OS/20170001P00353I0057.wav
+20170001P00085I0047 ST-CMDS-20170001_1-OS/20170001P00085I0047.wav
+20170001P00020I0017 ST-CMDS-20170001_1-OS/20170001P00020I0017.wav
+20170001P00120I0087 ST-CMDS-20170001_1-OS/20170001P00120I0087.wav
+20170001P00079I0087 ST-CMDS-20170001_1-OS/20170001P00079I0087.wav
+20170001P00329I0114 ST-CMDS-20170001_1-OS/20170001P00329I0114.wav
+20170001P00078A0003 ST-CMDS-20170001_1-OS/20170001P00078A0003.wav
+20170001P00189A0089 ST-CMDS-20170001_1-OS/20170001P00189A0089.wav
+20170001P00199A0096 ST-CMDS-20170001_1-OS/20170001P00199A0096.wav
+20170001P00183A0046 ST-CMDS-20170001_1-OS/20170001P00183A0046.wav
+20170001P00088A0044 ST-CMDS-20170001_1-OS/20170001P00088A0044.wav
+20170001P00049A0027 ST-CMDS-20170001_1-OS/20170001P00049A0027.wav
+20170001P00065I0104 ST-CMDS-20170001_1-OS/20170001P00065I0104.wav
+20170001P00351A0105 ST-CMDS-20170001_1-OS/20170001P00351A0105.wav
+20170001P00032A0033 ST-CMDS-20170001_1-OS/20170001P00032A0033.wav
+20170001P00117A0041 ST-CMDS-20170001_1-OS/20170001P00117A0041.wav
+20170001P00256A0119 ST-CMDS-20170001_1-OS/20170001P00256A0119.wav
+20170001P00332A0085 ST-CMDS-20170001_1-OS/20170001P00332A0085.wav
+20170001P00378I0018 ST-CMDS-20170001_1-OS/20170001P00378I0018.wav
+20170001P00363A0098 ST-CMDS-20170001_1-OS/20170001P00363A0098.wav
+20170001P00135A0017 ST-CMDS-20170001_1-OS/20170001P00135A0017.wav
+20170001P00384I0058 ST-CMDS-20170001_1-OS/20170001P00384I0058.wav
+20170001P00114A0072 ST-CMDS-20170001_1-OS/20170001P00114A0072.wav
+20170001P00119A0034 ST-CMDS-20170001_1-OS/20170001P00119A0034.wav
+20170001P00148I0096 ST-CMDS-20170001_1-OS/20170001P00148I0096.wav
+20170001P00163I0041 ST-CMDS-20170001_1-OS/20170001P00163I0041.wav
+20170001P00005I0116 ST-CMDS-20170001_1-OS/20170001P00005I0116.wav
+20170001P00349A0065 ST-CMDS-20170001_1-OS/20170001P00349A0065.wav
+20170001P00316I0061 ST-CMDS-20170001_1-OS/20170001P00316I0061.wav
+20170001P00007A0092 ST-CMDS-20170001_1-OS/20170001P00007A0092.wav
+20170001P00036I0112 ST-CMDS-20170001_1-OS/20170001P00036I0112.wav
+20170001P00172I0102 ST-CMDS-20170001_1-OS/20170001P00172I0102.wav
+20170001P00377I0038 ST-CMDS-20170001_1-OS/20170001P00377I0038.wav
+20170001P00347A0048 ST-CMDS-20170001_1-OS/20170001P00347A0048.wav
+20170001P00117A0022 ST-CMDS-20170001_1-OS/20170001P00117A0022.wav
+20170001P00269A0096 ST-CMDS-20170001_1-OS/20170001P00269A0096.wav
+20170001P00371I0083 ST-CMDS-20170001_1-OS/20170001P00371I0083.wav
+20170001P00200I0062 ST-CMDS-20170001_1-OS/20170001P00200I0062.wav
+20170001P00274I0105 ST-CMDS-20170001_1-OS/20170001P00274I0105.wav
+20170001P00209A0013 ST-CMDS-20170001_1-OS/20170001P00209A0013.wav
+20170001P00285I0030 ST-CMDS-20170001_1-OS/20170001P00285I0030.wav
+20170001P00018A0052 ST-CMDS-20170001_1-OS/20170001P00018A0052.wav
+20170001P00254I0087 ST-CMDS-20170001_1-OS/20170001P00254I0087.wav
+20170001P00334I0074 ST-CMDS-20170001_1-OS/20170001P00334I0074.wav
+20170001P00105I0055 ST-CMDS-20170001_1-OS/20170001P00105I0055.wav
+20170001P00057I0093 ST-CMDS-20170001_1-OS/20170001P00057I0093.wav
+20170001P00306I0069 ST-CMDS-20170001_1-OS/20170001P00306I0069.wav
+20170001P00214A0013 ST-CMDS-20170001_1-OS/20170001P00214A0013.wav
+20170001P00068A0082 ST-CMDS-20170001_1-OS/20170001P00068A0082.wav
+20170001P00263I0068 ST-CMDS-20170001_1-OS/20170001P00263I0068.wav
+20170001P00251A0083 ST-CMDS-20170001_1-OS/20170001P00251A0083.wav
+20170001P00263I0019 ST-CMDS-20170001_1-OS/20170001P00263I0019.wav
+20170001P00124I0029 ST-CMDS-20170001_1-OS/20170001P00124I0029.wav
+20170001P00428A0075 ST-CMDS-20170001_1-OS/20170001P00428A0075.wav
+20170001P00430A0042 ST-CMDS-20170001_1-OS/20170001P00430A0042.wav
+20170001P00431A0078 ST-CMDS-20170001_1-OS/20170001P00431A0078.wav
+20170001P00360A0015 ST-CMDS-20170001_1-OS/20170001P00360A0015.wav
+20170001P00430A0040 ST-CMDS-20170001_1-OS/20170001P00430A0040.wav
+20170001P00332I0105 ST-CMDS-20170001_1-OS/20170001P00332I0105.wav
+20170001P00274A0103 ST-CMDS-20170001_1-OS/20170001P00274A0103.wav
+20170001P00085I0085 ST-CMDS-20170001_1-OS/20170001P00085I0085.wav
+20170001P00383I0017 ST-CMDS-20170001_1-OS/20170001P00383I0017.wav
+20170001P00204I0117 ST-CMDS-20170001_1-OS/20170001P00204I0117.wav
+20170001P00222A0016 ST-CMDS-20170001_1-OS/20170001P00222A0016.wav
+20170001P00395I0117 ST-CMDS-20170001_1-OS/20170001P00395I0117.wav
+20170001P00430A0102 ST-CMDS-20170001_1-OS/20170001P00430A0102.wav
+20170001P00368I0039 ST-CMDS-20170001_1-OS/20170001P00368I0039.wav
+20170001P00088I0016 ST-CMDS-20170001_1-OS/20170001P00088I0016.wav
+20170001P00352I0024 ST-CMDS-20170001_1-OS/20170001P00352I0024.wav
+20170001P00143A0040 ST-CMDS-20170001_1-OS/20170001P00143A0040.wav
+20170001P00141A0107 ST-CMDS-20170001_1-OS/20170001P00141A0107.wav
+20170001P00043A0085 ST-CMDS-20170001_1-OS/20170001P00043A0085.wav
+20170001P00035I0105 ST-CMDS-20170001_1-OS/20170001P00035I0105.wav
+20170001P00325I0022 ST-CMDS-20170001_1-OS/20170001P00325I0022.wav
+20170001P00007A0080 ST-CMDS-20170001_1-OS/20170001P00007A0080.wav
+20170001P00406A0033 ST-CMDS-20170001_1-OS/20170001P00406A0033.wav
+20170001P00062A0108 ST-CMDS-20170001_1-OS/20170001P00062A0108.wav
+20170001P00280A0064 ST-CMDS-20170001_1-OS/20170001P00280A0064.wav
+20170001P00237I0106 ST-CMDS-20170001_1-OS/20170001P00237I0106.wav
+20170001P00217A0109 ST-CMDS-20170001_1-OS/20170001P00217A0109.wav
+20170001P00165A0106 ST-CMDS-20170001_1-OS/20170001P00165A0106.wav
+20170001P00030A0109 ST-CMDS-20170001_1-OS/20170001P00030A0109.wav
+20170001P00091A0043 ST-CMDS-20170001_1-OS/20170001P00091A0043.wav
+20170001P00032A0012 ST-CMDS-20170001_1-OS/20170001P00032A0012.wav
+20170001P00354I0108 ST-CMDS-20170001_1-OS/20170001P00354I0108.wav
+20170001P00107I0002 ST-CMDS-20170001_1-OS/20170001P00107I0002.wav
+20170001P00192I0062 ST-CMDS-20170001_1-OS/20170001P00192I0062.wav
+20170001P00277A0001 ST-CMDS-20170001_1-OS/20170001P00277A0001.wav
+20170001P00389I0103 ST-CMDS-20170001_1-OS/20170001P00389I0103.wav
+20170001P00354I0041 ST-CMDS-20170001_1-OS/20170001P00354I0041.wav
+20170001P00324I0108 ST-CMDS-20170001_1-OS/20170001P00324I0108.wav
+20170001P00116A0021 ST-CMDS-20170001_1-OS/20170001P00116A0021.wav
+20170001P00365I0019 ST-CMDS-20170001_1-OS/20170001P00365I0019.wav
+20170001P00175A0093 ST-CMDS-20170001_1-OS/20170001P00175A0093.wav
+20170001P00166I0071 ST-CMDS-20170001_1-OS/20170001P00166I0071.wav
+20170001P00025A0004 ST-CMDS-20170001_1-OS/20170001P00025A0004.wav
+20170001P00396A0077 ST-CMDS-20170001_1-OS/20170001P00396A0077.wav
+20170001P00073A0026 ST-CMDS-20170001_1-OS/20170001P00073A0026.wav
+20170001P00343I0112 ST-CMDS-20170001_1-OS/20170001P00343I0112.wav
+20170001P00352A0067 ST-CMDS-20170001_1-OS/20170001P00352A0067.wav
+20170001P00244I0108 ST-CMDS-20170001_1-OS/20170001P00244I0108.wav
+20170001P00317A0098 ST-CMDS-20170001_1-OS/20170001P00317A0098.wav
+20170001P00109A0022 ST-CMDS-20170001_1-OS/20170001P00109A0022.wav
+20170001P00315A0104 ST-CMDS-20170001_1-OS/20170001P00315A0104.wav
+20170001P00031I0018 ST-CMDS-20170001_1-OS/20170001P00031I0018.wav
+20170001P00050I0111 ST-CMDS-20170001_1-OS/20170001P00050I0111.wav
+20170001P00249A0002 ST-CMDS-20170001_1-OS/20170001P00249A0002.wav
+20170001P00268I0076 ST-CMDS-20170001_1-OS/20170001P00268I0076.wav
+20170001P00234A0020 ST-CMDS-20170001_1-OS/20170001P00234A0020.wav
+20170001P00205I0040 ST-CMDS-20170001_1-OS/20170001P00205I0040.wav
+20170001P00262I0089 ST-CMDS-20170001_1-OS/20170001P00262I0089.wav
+20170001P00201I0107 ST-CMDS-20170001_1-OS/20170001P00201I0107.wav
+20170001P00018I0056 ST-CMDS-20170001_1-OS/20170001P00018I0056.wav
+20170001P00028A0110 ST-CMDS-20170001_1-OS/20170001P00028A0110.wav
+20170001P00366I0018 ST-CMDS-20170001_1-OS/20170001P00366I0018.wav
+20170001P00065I0064 ST-CMDS-20170001_1-OS/20170001P00065I0064.wav
+20170001P00332I0053 ST-CMDS-20170001_1-OS/20170001P00332I0053.wav
+20170001P00002A0057 ST-CMDS-20170001_1-OS/20170001P00002A0057.wav
+20170001P00038A0001 ST-CMDS-20170001_1-OS/20170001P00038A0001.wav
+20170001P00364I0119 ST-CMDS-20170001_1-OS/20170001P00364I0119.wav
+20170001P00239I0067 ST-CMDS-20170001_1-OS/20170001P00239I0067.wav
+20170001P00339I0053 ST-CMDS-20170001_1-OS/20170001P00339I0053.wav
+20170001P00323I0116 ST-CMDS-20170001_1-OS/20170001P00323I0116.wav
+20170001P00144A0003 ST-CMDS-20170001_1-OS/20170001P00144A0003.wav
+20170001P00353I0052 ST-CMDS-20170001_1-OS/20170001P00353I0052.wav
+20170001P00010A0034 ST-CMDS-20170001_1-OS/20170001P00010A0034.wav
+20170001P00227I0072 ST-CMDS-20170001_1-OS/20170001P00227I0072.wav
+20170001P00397I0085 ST-CMDS-20170001_1-OS/20170001P00397I0085.wav
+20170001P00095A0016 ST-CMDS-20170001_1-OS/20170001P00095A0016.wav
+20170001P00360I0081 ST-CMDS-20170001_1-OS/20170001P00360I0081.wav
+20170001P00386A0078 ST-CMDS-20170001_1-OS/20170001P00386A0078.wav
+20170001P00409A0114 ST-CMDS-20170001_1-OS/20170001P00409A0114.wav
+20170001P00191I0022 ST-CMDS-20170001_1-OS/20170001P00191I0022.wav
+20170001P00041I0095 ST-CMDS-20170001_1-OS/20170001P00041I0095.wav
+20170001P00298A0007 ST-CMDS-20170001_1-OS/20170001P00298A0007.wav
+20170001P00176I0119 ST-CMDS-20170001_1-OS/20170001P00176I0119.wav
+20170001P00397I0042 ST-CMDS-20170001_1-OS/20170001P00397I0042.wav
+20170001P00376A0046 ST-CMDS-20170001_1-OS/20170001P00376A0046.wav
+20170001P00398A0031 ST-CMDS-20170001_1-OS/20170001P00398A0031.wav
+20170001P00089A0027 ST-CMDS-20170001_1-OS/20170001P00089A0027.wav
+20170001P00391A0015 ST-CMDS-20170001_1-OS/20170001P00391A0015.wav
+20170001P00243A0040 ST-CMDS-20170001_1-OS/20170001P00243A0040.wav
+20170001P00053I0119 ST-CMDS-20170001_1-OS/20170001P00053I0119.wav
+20170001P00193A0013 ST-CMDS-20170001_1-OS/20170001P00193A0013.wav
+20170001P00114A0051 ST-CMDS-20170001_1-OS/20170001P00114A0051.wav
+20170001P00293I0015 ST-CMDS-20170001_1-OS/20170001P00293I0015.wav
+20170001P00286I0044 ST-CMDS-20170001_1-OS/20170001P00286I0044.wav
+20170001P00191A0027 ST-CMDS-20170001_1-OS/20170001P00191A0027.wav
+20170001P00226I0084 ST-CMDS-20170001_1-OS/20170001P00226I0084.wav
+20170001P00116I0081 ST-CMDS-20170001_1-OS/20170001P00116I0081.wav
+20170001P00366A0017 ST-CMDS-20170001_1-OS/20170001P00366A0017.wav
+20170001P00106A0004 ST-CMDS-20170001_1-OS/20170001P00106A0004.wav
+20170001P00397A0007 ST-CMDS-20170001_1-OS/20170001P00397A0007.wav
+20170001P00422A0030 ST-CMDS-20170001_1-OS/20170001P00422A0030.wav
+20170001P00423A0111 ST-CMDS-20170001_1-OS/20170001P00423A0111.wav
+20170001P00288A0019 ST-CMDS-20170001_1-OS/20170001P00288A0019.wav
+20170001P00238A0096 ST-CMDS-20170001_1-OS/20170001P00238A0096.wav
+20170001P00330I0037 ST-CMDS-20170001_1-OS/20170001P00330I0037.wav
+20170001P00306I0068 ST-CMDS-20170001_1-OS/20170001P00306I0068.wav
+20170001P00399A0071 ST-CMDS-20170001_1-OS/20170001P00399A0071.wav
+20170001P00341A0061 ST-CMDS-20170001_1-OS/20170001P00341A0061.wav
+20170001P00309I0115 ST-CMDS-20170001_1-OS/20170001P00309I0115.wav
+20170001P00239A0086 ST-CMDS-20170001_1-OS/20170001P00239A0086.wav
+20170001P00133A0106 ST-CMDS-20170001_1-OS/20170001P00133A0106.wav
+20170001P00054A0106 ST-CMDS-20170001_1-OS/20170001P00054A0106.wav
+20170001P00117I0080 ST-CMDS-20170001_1-OS/20170001P00117I0080.wav
+20170001P00233A0026 ST-CMDS-20170001_1-OS/20170001P00233A0026.wav
+20170001P00347I0113 ST-CMDS-20170001_1-OS/20170001P00347I0113.wav
+20170001P00074I0028 ST-CMDS-20170001_1-OS/20170001P00074I0028.wav
+20170001P00390A0031 ST-CMDS-20170001_1-OS/20170001P00390A0031.wav
+20170001P00390A0050 ST-CMDS-20170001_1-OS/20170001P00390A0050.wav
+20170001P00224A0089 ST-CMDS-20170001_1-OS/20170001P00224A0089.wav
+20170001P00012I0097 ST-CMDS-20170001_1-OS/20170001P00012I0097.wav
+20170001P00177I0065 ST-CMDS-20170001_1-OS/20170001P00177I0065.wav
+20170001P00241A0092 ST-CMDS-20170001_1-OS/20170001P00241A0092.wav
+20170001P00290I0027 ST-CMDS-20170001_1-OS/20170001P00290I0027.wav
+20170001P00300I0094 ST-CMDS-20170001_1-OS/20170001P00300I0094.wav
+20170001P00075A0117 ST-CMDS-20170001_1-OS/20170001P00075A0117.wav
+20170001P00010I0080 ST-CMDS-20170001_1-OS/20170001P00010I0080.wav
+20170001P00108A0069 ST-CMDS-20170001_1-OS/20170001P00108A0069.wav
+20170001P00185A0059 ST-CMDS-20170001_1-OS/20170001P00185A0059.wav
+20170001P00404A0075 ST-CMDS-20170001_1-OS/20170001P00404A0075.wav
+20170001P00120I0040 ST-CMDS-20170001_1-OS/20170001P00120I0040.wav
+20170001P00330I0068 ST-CMDS-20170001_1-OS/20170001P00330I0068.wav
+20170001P00311A0046 ST-CMDS-20170001_1-OS/20170001P00311A0046.wav
+20170001P00021I0012 ST-CMDS-20170001_1-OS/20170001P00021I0012.wav
+20170001P00177A0099 ST-CMDS-20170001_1-OS/20170001P00177A0099.wav
+20170001P00138I0002 ST-CMDS-20170001_1-OS/20170001P00138I0002.wav
+20170001P00432A0103 ST-CMDS-20170001_1-OS/20170001P00432A0103.wav
+20170001P00101A0048 ST-CMDS-20170001_1-OS/20170001P00101A0048.wav
+20170001P00344A0065 ST-CMDS-20170001_1-OS/20170001P00344A0065.wav
+20170001P00442A0019 ST-CMDS-20170001_1-OS/20170001P00442A0019.wav
+20170001P00148I0047 ST-CMDS-20170001_1-OS/20170001P00148I0047.wav
+20170001P00325A0055 ST-CMDS-20170001_1-OS/20170001P00325A0055.wav
+20170001P00172I0058 ST-CMDS-20170001_1-OS/20170001P00172I0058.wav
+20170001P00026I0116 ST-CMDS-20170001_1-OS/20170001P00026I0116.wav
+20170001P00334I0031 ST-CMDS-20170001_1-OS/20170001P00334I0031.wav
+20170001P00343I0058 ST-CMDS-20170001_1-OS/20170001P00343I0058.wav
+20170001P00063I0051 ST-CMDS-20170001_1-OS/20170001P00063I0051.wav
+20170001P00080A0109 ST-CMDS-20170001_1-OS/20170001P00080A0109.wav
+20170001P00166I0023 ST-CMDS-20170001_1-OS/20170001P00166I0023.wav
+20170001P00200A0112 ST-CMDS-20170001_1-OS/20170001P00200A0112.wav
+20170001P00258I0099 ST-CMDS-20170001_1-OS/20170001P00258I0099.wav
+20170001P00424A0041 ST-CMDS-20170001_1-OS/20170001P00424A0041.wav
+20170001P00211A0085 ST-CMDS-20170001_1-OS/20170001P00211A0085.wav
+20170001P00364A0094 ST-CMDS-20170001_1-OS/20170001P00364A0094.wav
+20170001P00091I0082 ST-CMDS-20170001_1-OS/20170001P00091I0082.wav
+20170001P00277A0100 ST-CMDS-20170001_1-OS/20170001P00277A0100.wav
+20170001P00259I0080 ST-CMDS-20170001_1-OS/20170001P00259I0080.wav
+20170001P00189I0018 ST-CMDS-20170001_1-OS/20170001P00189I0018.wav
+20170001P00204I0068 ST-CMDS-20170001_1-OS/20170001P00204I0068.wav
+20170001P00198A0036 ST-CMDS-20170001_1-OS/20170001P00198A0036.wav
+20170001P00405I0114 ST-CMDS-20170001_1-OS/20170001P00405I0114.wav
+20170001P00135A0052 ST-CMDS-20170001_1-OS/20170001P00135A0052.wav
+20170001P00364I0035 ST-CMDS-20170001_1-OS/20170001P00364I0035.wav
+20170001P00133I0062 ST-CMDS-20170001_1-OS/20170001P00133I0062.wav
+20170001P00103A0109 ST-CMDS-20170001_1-OS/20170001P00103A0109.wav
+20170001P00018I0033 ST-CMDS-20170001_1-OS/20170001P00018I0033.wav
+20170001P00304I0092 ST-CMDS-20170001_1-OS/20170001P00304I0092.wav
+20170001P00325A0009 ST-CMDS-20170001_1-OS/20170001P00325A0009.wav
+20170001P00358I0079 ST-CMDS-20170001_1-OS/20170001P00358I0079.wav
+20170001P00379A0024 ST-CMDS-20170001_1-OS/20170001P00379A0024.wav
+20170001P00265A0048 ST-CMDS-20170001_1-OS/20170001P00265A0048.wav
+20170001P00433A0078 ST-CMDS-20170001_1-OS/20170001P00433A0078.wav
+20170001P00245A0019 ST-CMDS-20170001_1-OS/20170001P00245A0019.wav
+20170001P00201A0118 ST-CMDS-20170001_1-OS/20170001P00201A0118.wav
+20170001P00248I0055 ST-CMDS-20170001_1-OS/20170001P00248I0055.wav
+20170001P00243I0046 ST-CMDS-20170001_1-OS/20170001P00243I0046.wav
+20170001P00002I0035 ST-CMDS-20170001_1-OS/20170001P00002I0035.wav
+20170001P00367I0075 ST-CMDS-20170001_1-OS/20170001P00367I0075.wav
+20170001P00019A0067 ST-CMDS-20170001_1-OS/20170001P00019A0067.wav
+20170001P00359A0115 ST-CMDS-20170001_1-OS/20170001P00359A0115.wav
+20170001P00326A0098 ST-CMDS-20170001_1-OS/20170001P00326A0098.wav
+20170001P00326I0038 ST-CMDS-20170001_1-OS/20170001P00326I0038.wav
+20170001P00085I0111 ST-CMDS-20170001_1-OS/20170001P00085I0111.wav
+20170001P00201A0008 ST-CMDS-20170001_1-OS/20170001P00201A0008.wav
+20170001P00133I0068 ST-CMDS-20170001_1-OS/20170001P00133I0068.wav
+20170001P00161A0021 ST-CMDS-20170001_1-OS/20170001P00161A0021.wav
+20170001P00028A0087 ST-CMDS-20170001_1-OS/20170001P00028A0087.wav
+20170001P00414A0074 ST-CMDS-20170001_1-OS/20170001P00414A0074.wav
+20170001P00204A0108 ST-CMDS-20170001_1-OS/20170001P00204A0108.wav
+20170001P00342A0020 ST-CMDS-20170001_1-OS/20170001P00342A0020.wav
+20170001P00378A0119 ST-CMDS-20170001_1-OS/20170001P00378A0119.wav
+20170001P00356A0061 ST-CMDS-20170001_1-OS/20170001P00356A0061.wav
+20170001P00082A0043 ST-CMDS-20170001_1-OS/20170001P00082A0043.wav
+20170001P00215A0061 ST-CMDS-20170001_1-OS/20170001P00215A0061.wav
+20170001P00302A0082 ST-CMDS-20170001_1-OS/20170001P00302A0082.wav
+20170001P00065I0111 ST-CMDS-20170001_1-OS/20170001P00065I0111.wav
+20170001P00025I0088 ST-CMDS-20170001_1-OS/20170001P00025I0088.wav
+20170001P00373A0105 ST-CMDS-20170001_1-OS/20170001P00373A0105.wav
+20170001P00298I0063 ST-CMDS-20170001_1-OS/20170001P00298I0063.wav
+20170001P00256I0065 ST-CMDS-20170001_1-OS/20170001P00256I0065.wav
+20170001P00089I0026 ST-CMDS-20170001_1-OS/20170001P00089I0026.wav
+20170001P00162I0109 ST-CMDS-20170001_1-OS/20170001P00162I0109.wav
+20170001P00119A0007 ST-CMDS-20170001_1-OS/20170001P00119A0007.wav
+20170001P00178I0031 ST-CMDS-20170001_1-OS/20170001P00178I0031.wav
+20170001P00082I0057 ST-CMDS-20170001_1-OS/20170001P00082I0057.wav
+20170001P00122I0054 ST-CMDS-20170001_1-OS/20170001P00122I0054.wav
+20170001P00425A0053 ST-CMDS-20170001_1-OS/20170001P00425A0053.wav
+20170001P00019A0075 ST-CMDS-20170001_1-OS/20170001P00019A0075.wav
+20170001P00245I0106 ST-CMDS-20170001_1-OS/20170001P00245I0106.wav
+20170001P00351I0069 ST-CMDS-20170001_1-OS/20170001P00351I0069.wav
+20170001P00110I0051 ST-CMDS-20170001_1-OS/20170001P00110I0051.wav
+20170001P00196A0099 ST-CMDS-20170001_1-OS/20170001P00196A0099.wav
+20170001P00128A0062 ST-CMDS-20170001_1-OS/20170001P00128A0062.wav
+20170001P00218A0056 ST-CMDS-20170001_1-OS/20170001P00218A0056.wav
+20170001P00406A0085 ST-CMDS-20170001_1-OS/20170001P00406A0085.wav
+20170001P00170A0002 ST-CMDS-20170001_1-OS/20170001P00170A0002.wav
+20170001P00139I0089 ST-CMDS-20170001_1-OS/20170001P00139I0089.wav
+20170001P00290A0038 ST-CMDS-20170001_1-OS/20170001P00290A0038.wav
+20170001P00230A0097 ST-CMDS-20170001_1-OS/20170001P00230A0097.wav
+20170001P00300I0046 ST-CMDS-20170001_1-OS/20170001P00300I0046.wav
+20170001P00157I0015 ST-CMDS-20170001_1-OS/20170001P00157I0015.wav
+20170001P00030I0098 ST-CMDS-20170001_1-OS/20170001P00030I0098.wav
+20170001P00152A0107 ST-CMDS-20170001_1-OS/20170001P00152A0107.wav
+20170001P00049I0099 ST-CMDS-20170001_1-OS/20170001P00049I0099.wav
+20170001P00287I0061 ST-CMDS-20170001_1-OS/20170001P00287I0061.wav
+20170001P00284A0003 ST-CMDS-20170001_1-OS/20170001P00284A0003.wav
+20170001P00359A0015 ST-CMDS-20170001_1-OS/20170001P00359A0015.wav
+20170001P00269A0045 ST-CMDS-20170001_1-OS/20170001P00269A0045.wav
+20170001P00385A0066 ST-CMDS-20170001_1-OS/20170001P00385A0066.wav
+20170001P00211I0064 ST-CMDS-20170001_1-OS/20170001P00211I0064.wav
+20170001P00127I0004 ST-CMDS-20170001_1-OS/20170001P00127I0004.wav
+20170001P00271A0043 ST-CMDS-20170001_1-OS/20170001P00271A0043.wav
+20170001P00220I0085 ST-CMDS-20170001_1-OS/20170001P00220I0085.wav
+20170001P00215A0015 ST-CMDS-20170001_1-OS/20170001P00215A0015.wav
+20170001P00093I0119 ST-CMDS-20170001_1-OS/20170001P00093I0119.wav
+20170001P00424A0092 ST-CMDS-20170001_1-OS/20170001P00424A0092.wav
+20170001P00214I0118 ST-CMDS-20170001_1-OS/20170001P00214I0118.wav
+20170001P00178A0079 ST-CMDS-20170001_1-OS/20170001P00178A0079.wav
+20170001P00082A0046 ST-CMDS-20170001_1-OS/20170001P00082A0046.wav
+20170001P00369A0062 ST-CMDS-20170001_1-OS/20170001P00369A0062.wav
+20170001P00016I0085 ST-CMDS-20170001_1-OS/20170001P00016I0085.wav
+20170001P00062I0043 ST-CMDS-20170001_1-OS/20170001P00062I0043.wav
+20170001P00180A0005 ST-CMDS-20170001_1-OS/20170001P00180A0005.wav
+20170001P00228A0083 ST-CMDS-20170001_1-OS/20170001P00228A0083.wav
+20170001P00058A0066 ST-CMDS-20170001_1-OS/20170001P00058A0066.wav
+20170001P00282A0112 ST-CMDS-20170001_1-OS/20170001P00282A0112.wav
+20170001P00402I0023 ST-CMDS-20170001_1-OS/20170001P00402I0023.wav
+20170001P00175A0051 ST-CMDS-20170001_1-OS/20170001P00175A0051.wav
+20170001P00383A0006 ST-CMDS-20170001_1-OS/20170001P00383A0006.wav
+20170001P00280A0029 ST-CMDS-20170001_1-OS/20170001P00280A0029.wav
+20170001P00422A0010 ST-CMDS-20170001_1-OS/20170001P00422A0010.wav
+20170001P00433A0011 ST-CMDS-20170001_1-OS/20170001P00433A0011.wav
+20170001P00064A0058 ST-CMDS-20170001_1-OS/20170001P00064A0058.wav
+20170001P00392A0109 ST-CMDS-20170001_1-OS/20170001P00392A0109.wav
+20170001P00193I0008 ST-CMDS-20170001_1-OS/20170001P00193I0008.wav
+20170001P00177A0117 ST-CMDS-20170001_1-OS/20170001P00177A0117.wav
+20170001P00004I0078 ST-CMDS-20170001_1-OS/20170001P00004I0078.wav
+20170001P00399A0111 ST-CMDS-20170001_1-OS/20170001P00399A0111.wav
+20170001P00075I0007 ST-CMDS-20170001_1-OS/20170001P00075I0007.wav
+20170001P00355I0002 ST-CMDS-20170001_1-OS/20170001P00355I0002.wav
+20170001P00012A0004 ST-CMDS-20170001_1-OS/20170001P00012A0004.wav
+20170001P00164I0034 ST-CMDS-20170001_1-OS/20170001P00164I0034.wav
+20170001P00333I0049 ST-CMDS-20170001_1-OS/20170001P00333I0049.wav
+20170001P00403I0081 ST-CMDS-20170001_1-OS/20170001P00403I0081.wav
+20170001P00086I0047 ST-CMDS-20170001_1-OS/20170001P00086I0047.wav
+20170001P00076I0106 ST-CMDS-20170001_1-OS/20170001P00076I0106.wav
+20170001P00368I0043 ST-CMDS-20170001_1-OS/20170001P00368I0043.wav
+20170001P00385I0040 ST-CMDS-20170001_1-OS/20170001P00385I0040.wav
+20170001P00119A0109 ST-CMDS-20170001_1-OS/20170001P00119A0109.wav
+20170001P00062A0101 ST-CMDS-20170001_1-OS/20170001P00062A0101.wav
+20170001P00226I0010 ST-CMDS-20170001_1-OS/20170001P00226I0010.wav
+20170001P00353A0011 ST-CMDS-20170001_1-OS/20170001P00353A0011.wav
+20170001P00198A0007 ST-CMDS-20170001_1-OS/20170001P00198A0007.wav
+20170001P00284A0034 ST-CMDS-20170001_1-OS/20170001P00284A0034.wav
+20170001P00080A0039 ST-CMDS-20170001_1-OS/20170001P00080A0039.wav
+20170001P00120A0004 ST-CMDS-20170001_1-OS/20170001P00120A0004.wav
+20170001P00381A0079 ST-CMDS-20170001_1-OS/20170001P00381A0079.wav
+20170001P00396I0100 ST-CMDS-20170001_1-OS/20170001P00396I0100.wav
+20170001P00272A0095 ST-CMDS-20170001_1-OS/20170001P00272A0095.wav
+20170001P00126A0025 ST-CMDS-20170001_1-OS/20170001P00126A0025.wav
+20170001P00091I0096 ST-CMDS-20170001_1-OS/20170001P00091I0096.wav
+20170001P00190I0085 ST-CMDS-20170001_1-OS/20170001P00190I0085.wav
+20170001P00314I0016 ST-CMDS-20170001_1-OS/20170001P00314I0016.wav
+20170001P00394A0087 ST-CMDS-20170001_1-OS/20170001P00394A0087.wav
+20170001P00103I0043 ST-CMDS-20170001_1-OS/20170001P00103I0043.wav
+20170001P00279A0078 ST-CMDS-20170001_1-OS/20170001P00279A0078.wav
+20170001P00217I0011 ST-CMDS-20170001_1-OS/20170001P00217I0011.wav
+20170001P00188A0024 ST-CMDS-20170001_1-OS/20170001P00188A0024.wav
+20170001P00073A0104 ST-CMDS-20170001_1-OS/20170001P00073A0104.wav
+20170001P00372I0108 ST-CMDS-20170001_1-OS/20170001P00372I0108.wav
+20170001P00230A0039 ST-CMDS-20170001_1-OS/20170001P00230A0039.wav
+20170001P00177A0100 ST-CMDS-20170001_1-OS/20170001P00177A0100.wav
+20170001P00395A0061 ST-CMDS-20170001_1-OS/20170001P00395A0061.wav
+20170001P00415A0079 ST-CMDS-20170001_1-OS/20170001P00415A0079.wav
+20170001P00074A0097 ST-CMDS-20170001_1-OS/20170001P00074A0097.wav
+20170001P00235I0089 ST-CMDS-20170001_1-OS/20170001P00235I0089.wav
+20170001P00132A0079 ST-CMDS-20170001_1-OS/20170001P00132A0079.wav
+20170001P00038I0035 ST-CMDS-20170001_1-OS/20170001P00038I0035.wav
+20170001P00393I0050 ST-CMDS-20170001_1-OS/20170001P00393I0050.wav
+20170001P00225A0086 ST-CMDS-20170001_1-OS/20170001P00225A0086.wav
+20170001P00144A0106 ST-CMDS-20170001_1-OS/20170001P00144A0106.wav
+20170001P00150A0112 ST-CMDS-20170001_1-OS/20170001P00150A0112.wav
+20170001P00159I0051 ST-CMDS-20170001_1-OS/20170001P00159I0051.wav
+20170001P00419A0063 ST-CMDS-20170001_1-OS/20170001P00419A0063.wav
+20170001P00303I0106 ST-CMDS-20170001_1-OS/20170001P00303I0106.wav
+20170001P00195A0022 ST-CMDS-20170001_1-OS/20170001P00195A0022.wav
+20170001P00304A0011 ST-CMDS-20170001_1-OS/20170001P00304A0011.wav
+20170001P00238I0113 ST-CMDS-20170001_1-OS/20170001P00238I0113.wav
+20170001P00395I0007 ST-CMDS-20170001_1-OS/20170001P00395I0007.wav
+20170001P00333A0039 ST-CMDS-20170001_1-OS/20170001P00333A0039.wav
+20170001P00062I0018 ST-CMDS-20170001_1-OS/20170001P00062I0018.wav
+20170001P00098I0083 ST-CMDS-20170001_1-OS/20170001P00098I0083.wav
+20170001P00239A0077 ST-CMDS-20170001_1-OS/20170001P00239A0077.wav
+20170001P00393I0086 ST-CMDS-20170001_1-OS/20170001P00393I0086.wav
+20170001P00167I0008 ST-CMDS-20170001_1-OS/20170001P00167I0008.wav
+20170001P00107A0042 ST-CMDS-20170001_1-OS/20170001P00107A0042.wav
+20170001P00159A0081 ST-CMDS-20170001_1-OS/20170001P00159A0081.wav
+20170001P00006I0064 ST-CMDS-20170001_1-OS/20170001P00006I0064.wav
+20170001P00229I0021 ST-CMDS-20170001_1-OS/20170001P00229I0021.wav
+20170001P00135I0052 ST-CMDS-20170001_1-OS/20170001P00135I0052.wav
+20170001P00160I0034 ST-CMDS-20170001_1-OS/20170001P00160I0034.wav
+20170001P00408I0053 ST-CMDS-20170001_1-OS/20170001P00408I0053.wav
+20170001P00391I0114 ST-CMDS-20170001_1-OS/20170001P00391I0114.wav
+20170001P00024I0056 ST-CMDS-20170001_1-OS/20170001P00024I0056.wav
+20170001P00277A0029 ST-CMDS-20170001_1-OS/20170001P00277A0029.wav
+20170001P00356I0020 ST-CMDS-20170001_1-OS/20170001P00356I0020.wav
+20170001P00021I0073 ST-CMDS-20170001_1-OS/20170001P00021I0073.wav
+20170001P00076A0027 ST-CMDS-20170001_1-OS/20170001P00076A0027.wav
+20170001P00148I0091 ST-CMDS-20170001_1-OS/20170001P00148I0091.wav
+20170001P00153A0033 ST-CMDS-20170001_1-OS/20170001P00153A0033.wav
+20170001P00163I0008 ST-CMDS-20170001_1-OS/20170001P00163I0008.wav
+20170001P00354A0014 ST-CMDS-20170001_1-OS/20170001P00354A0014.wav
+20170001P00146I0081 ST-CMDS-20170001_1-OS/20170001P00146I0081.wav
+20170001P00434A0071 ST-CMDS-20170001_1-OS/20170001P00434A0071.wav
+20170001P00426A0080 ST-CMDS-20170001_1-OS/20170001P00426A0080.wav
+20170001P00366A0058 ST-CMDS-20170001_1-OS/20170001P00366A0058.wav
+20170001P00181A0103 ST-CMDS-20170001_1-OS/20170001P00181A0103.wav
+20170001P00242I0095 ST-CMDS-20170001_1-OS/20170001P00242I0095.wav
+20170001P00107I0094 ST-CMDS-20170001_1-OS/20170001P00107I0094.wav
+20170001P00299A0032 ST-CMDS-20170001_1-OS/20170001P00299A0032.wav
+20170001P00282A0088 ST-CMDS-20170001_1-OS/20170001P00282A0088.wav
+20170001P00304A0008 ST-CMDS-20170001_1-OS/20170001P00304A0008.wav
+20170001P00289A0084 ST-CMDS-20170001_1-OS/20170001P00289A0084.wav
+20170001P00139I0009 ST-CMDS-20170001_1-OS/20170001P00139I0009.wav
+20170001P00289A0053 ST-CMDS-20170001_1-OS/20170001P00289A0053.wav
+20170001P00141I0079 ST-CMDS-20170001_1-OS/20170001P00141I0079.wav
+20170001P00138I0033 ST-CMDS-20170001_1-OS/20170001P00138I0033.wav
+20170001P00318A0066 ST-CMDS-20170001_1-OS/20170001P00318A0066.wav
+20170001P00263I0088 ST-CMDS-20170001_1-OS/20170001P00263I0088.wav
+20170001P00226I0083 ST-CMDS-20170001_1-OS/20170001P00226I0083.wav
+20170001P00094I0101 ST-CMDS-20170001_1-OS/20170001P00094I0101.wav
+20170001P00039A0108 ST-CMDS-20170001_1-OS/20170001P00039A0108.wav
+20170001P00103I0045 ST-CMDS-20170001_1-OS/20170001P00103I0045.wav
+20170001P00379I0055 ST-CMDS-20170001_1-OS/20170001P00379I0055.wav
+20170001P00412A0037 ST-CMDS-20170001_1-OS/20170001P00412A0037.wav
+20170001P00143I0092 ST-CMDS-20170001_1-OS/20170001P00143I0092.wav
+20170001P00252A0113 ST-CMDS-20170001_1-OS/20170001P00252A0113.wav
+20170001P00319I0114 ST-CMDS-20170001_1-OS/20170001P00319I0114.wav
+20170001P00425A0007 ST-CMDS-20170001_1-OS/20170001P00425A0007.wav
+20170001P00278I0038 ST-CMDS-20170001_1-OS/20170001P00278I0038.wav
+20170001P00042I0086 ST-CMDS-20170001_1-OS/20170001P00042I0086.wav
+20170001P00028A0016 ST-CMDS-20170001_1-OS/20170001P00028A0016.wav
+20170001P00326A0051 ST-CMDS-20170001_1-OS/20170001P00326A0051.wav
+20170001P00101A0084 ST-CMDS-20170001_1-OS/20170001P00101A0084.wav
+20170001P00241A0015 ST-CMDS-20170001_1-OS/20170001P00241A0015.wav
+20170001P00023A0047 ST-CMDS-20170001_1-OS/20170001P00023A0047.wav
+20170001P00164A0084 ST-CMDS-20170001_1-OS/20170001P00164A0084.wav
+20170001P00360I0004 ST-CMDS-20170001_1-OS/20170001P00360I0004.wav
+20170001P00381I0096 ST-CMDS-20170001_1-OS/20170001P00381I0096.wav
+20170001P00310I0021 ST-CMDS-20170001_1-OS/20170001P00310I0021.wav
+20170001P00184A0037 ST-CMDS-20170001_1-OS/20170001P00184A0037.wav
+20170001P00212A0074 ST-CMDS-20170001_1-OS/20170001P00212A0074.wav
+20170001P00362A0077 ST-CMDS-20170001_1-OS/20170001P00362A0077.wav
+20170001P00157A0105 ST-CMDS-20170001_1-OS/20170001P00157A0105.wav
+20170001P00166I0004 ST-CMDS-20170001_1-OS/20170001P00166I0004.wav
+20170001P00375A0042 ST-CMDS-20170001_1-OS/20170001P00375A0042.wav
+20170001P00002I0013 ST-CMDS-20170001_1-OS/20170001P00002I0013.wav
+20170001P00081I0053 ST-CMDS-20170001_1-OS/20170001P00081I0053.wav
+20170001P00130I0075 ST-CMDS-20170001_1-OS/20170001P00130I0075.wav
+20170001P00022I0037 ST-CMDS-20170001_1-OS/20170001P00022I0037.wav
+20170001P00045I0052 ST-CMDS-20170001_1-OS/20170001P00045I0052.wav
+20170001P00347I0117 ST-CMDS-20170001_1-OS/20170001P00347I0117.wav
+20170001P00177I0086 ST-CMDS-20170001_1-OS/20170001P00177I0086.wav
+20170001P00067A0009 ST-CMDS-20170001_1-OS/20170001P00067A0009.wav
+20170001P00101I0104 ST-CMDS-20170001_1-OS/20170001P00101I0104.wav
+20170001P00402A0053 ST-CMDS-20170001_1-OS/20170001P00402A0053.wav
+20170001P00382I0056 ST-CMDS-20170001_1-OS/20170001P00382I0056.wav
+20170001P00197I0042 ST-CMDS-20170001_1-OS/20170001P00197I0042.wav
+20170001P00151A0119 ST-CMDS-20170001_1-OS/20170001P00151A0119.wav
+20170001P00176I0117 ST-CMDS-20170001_1-OS/20170001P00176I0117.wav
+20170001P00053I0061 ST-CMDS-20170001_1-OS/20170001P00053I0061.wav
+20170001P00338A0023 ST-CMDS-20170001_1-OS/20170001P00338A0023.wav
+20170001P00095A0004 ST-CMDS-20170001_1-OS/20170001P00095A0004.wav
+20170001P00241A0114 ST-CMDS-20170001_1-OS/20170001P00241A0114.wav
+20170001P00144I0076 ST-CMDS-20170001_1-OS/20170001P00144I0076.wav
+20170001P00036A0096 ST-CMDS-20170001_1-OS/20170001P00036A0096.wav
+20170001P00018I0070 ST-CMDS-20170001_1-OS/20170001P00018I0070.wav
+20170001P00084A0103 ST-CMDS-20170001_1-OS/20170001P00084A0103.wav
+20170001P00385A0020 ST-CMDS-20170001_1-OS/20170001P00385A0020.wav
+20170001P00399I0014 ST-CMDS-20170001_1-OS/20170001P00399I0014.wav
+20170001P00137A0047 ST-CMDS-20170001_1-OS/20170001P00137A0047.wav
+20170001P00393I0030 ST-CMDS-20170001_1-OS/20170001P00393I0030.wav
+20170001P00103A0011 ST-CMDS-20170001_1-OS/20170001P00103A0011.wav
+20170001P00383A0030 ST-CMDS-20170001_1-OS/20170001P00383A0030.wav
+20170001P00083I0075 ST-CMDS-20170001_1-OS/20170001P00083I0075.wav
+20170001P00299I0017 ST-CMDS-20170001_1-OS/20170001P00299I0017.wav
+20170001P00277I0071 ST-CMDS-20170001_1-OS/20170001P00277I0071.wav
+20170001P00128I0006 ST-CMDS-20170001_1-OS/20170001P00128I0006.wav
+20170001P00256I0009 ST-CMDS-20170001_1-OS/20170001P00256I0009.wav
+20170001P00166I0069 ST-CMDS-20170001_1-OS/20170001P00166I0069.wav
+20170001P00226A0013 ST-CMDS-20170001_1-OS/20170001P00226A0013.wav
+20170001P00080A0057 ST-CMDS-20170001_1-OS/20170001P00080A0057.wav
+20170001P00236I0039 ST-CMDS-20170001_1-OS/20170001P00236I0039.wav
+20170001P00068A0041 ST-CMDS-20170001_1-OS/20170001P00068A0041.wav
+20170001P00202A0094 ST-CMDS-20170001_1-OS/20170001P00202A0094.wav
+20170001P00215A0070 ST-CMDS-20170001_1-OS/20170001P00215A0070.wav
+20170001P00253I0107 ST-CMDS-20170001_1-OS/20170001P00253I0107.wav
+20170001P00024A0046 ST-CMDS-20170001_1-OS/20170001P00024A0046.wav
+20170001P00299A0075 ST-CMDS-20170001_1-OS/20170001P00299A0075.wav
+20170001P00295I0035 ST-CMDS-20170001_1-OS/20170001P00295I0035.wav
+20170001P00374A0068 ST-CMDS-20170001_1-OS/20170001P00374A0068.wav
+20170001P00412A0094 ST-CMDS-20170001_1-OS/20170001P00412A0094.wav
+20170001P00208I0019 ST-CMDS-20170001_1-OS/20170001P00208I0019.wav
+20170001P00185A0002 ST-CMDS-20170001_1-OS/20170001P00185A0002.wav
+20170001P00169A0095 ST-CMDS-20170001_1-OS/20170001P00169A0095.wav
+20170001P00271I0065 ST-CMDS-20170001_1-OS/20170001P00271I0065.wav
+20170001P00287A0102 ST-CMDS-20170001_1-OS/20170001P00287A0102.wav
+20170001P00126I0063 ST-CMDS-20170001_1-OS/20170001P00126I0063.wav
+20170001P00218I0018 ST-CMDS-20170001_1-OS/20170001P00218I0018.wav
+20170001P00048A0007 ST-CMDS-20170001_1-OS/20170001P00048A0007.wav
+20170001P00390A0091 ST-CMDS-20170001_1-OS/20170001P00390A0091.wav
+20170001P00129A0077 ST-CMDS-20170001_1-OS/20170001P00129A0077.wav
+20170001P00115I0001 ST-CMDS-20170001_1-OS/20170001P00115I0001.wav
+20170001P00039I0103 ST-CMDS-20170001_1-OS/20170001P00039I0103.wav
+20170001P00278A0003 ST-CMDS-20170001_1-OS/20170001P00278A0003.wav
+20170001P00092A0068 ST-CMDS-20170001_1-OS/20170001P00092A0068.wav
+20170001P00160A0119 ST-CMDS-20170001_1-OS/20170001P00160A0119.wav
+20170001P00158I0069 ST-CMDS-20170001_1-OS/20170001P00158I0069.wav
+20170001P00076A0082 ST-CMDS-20170001_1-OS/20170001P00076A0082.wav
+20170001P00010I0106 ST-CMDS-20170001_1-OS/20170001P00010I0106.wav
+20170001P00093I0053 ST-CMDS-20170001_1-OS/20170001P00093I0053.wav
+20170001P00412A0082 ST-CMDS-20170001_1-OS/20170001P00412A0082.wav
+20170001P00410I0116 ST-CMDS-20170001_1-OS/20170001P00410I0116.wav
+20170001P00218A0068 ST-CMDS-20170001_1-OS/20170001P00218A0068.wav
+20170001P00074I0007 ST-CMDS-20170001_1-OS/20170001P00074I0007.wav
+20170001P00026A0022 ST-CMDS-20170001_1-OS/20170001P00026A0022.wav
+20170001P00235I0052 ST-CMDS-20170001_1-OS/20170001P00235I0052.wav
+20170001P00331I0098 ST-CMDS-20170001_1-OS/20170001P00331I0098.wav
+20170001P00226I0033 ST-CMDS-20170001_1-OS/20170001P00226I0033.wav
+20170001P00055A0081 ST-CMDS-20170001_1-OS/20170001P00055A0081.wav
+20170001P00233I0018 ST-CMDS-20170001_1-OS/20170001P00233I0018.wav
+20170001P00127A0061 ST-CMDS-20170001_1-OS/20170001P00127A0061.wav
+20170001P00037I0086 ST-CMDS-20170001_1-OS/20170001P00037I0086.wav
+20170001P00442A0041 ST-CMDS-20170001_1-OS/20170001P00442A0041.wav
+20170001P00205I0118 ST-CMDS-20170001_1-OS/20170001P00205I0118.wav
+20170001P00052I0059 ST-CMDS-20170001_1-OS/20170001P00052I0059.wav
+20170001P00071A0100 ST-CMDS-20170001_1-OS/20170001P00071A0100.wav
+20170001P00148A0113 ST-CMDS-20170001_1-OS/20170001P00148A0113.wav
+20170001P00319I0049 ST-CMDS-20170001_1-OS/20170001P00319I0049.wav
+20170001P00045I0047 ST-CMDS-20170001_1-OS/20170001P00045I0047.wav
+20170001P00303A0068 ST-CMDS-20170001_1-OS/20170001P00303A0068.wav
+20170001P00126I0024 ST-CMDS-20170001_1-OS/20170001P00126I0024.wav
+20170001P00254I0031 ST-CMDS-20170001_1-OS/20170001P00254I0031.wav
+20170001P00086A0021 ST-CMDS-20170001_1-OS/20170001P00086A0021.wav
+20170001P00038I0059 ST-CMDS-20170001_1-OS/20170001P00038I0059.wav
+20170001P00030A0104 ST-CMDS-20170001_1-OS/20170001P00030A0104.wav
+20170001P00075A0026 ST-CMDS-20170001_1-OS/20170001P00075A0026.wav
+20170001P00069A0118 ST-CMDS-20170001_1-OS/20170001P00069A0118.wav
+20170001P00272I0060 ST-CMDS-20170001_1-OS/20170001P00272I0060.wav
+20170001P00035A0003 ST-CMDS-20170001_1-OS/20170001P00035A0003.wav
+20170001P00175A0074 ST-CMDS-20170001_1-OS/20170001P00175A0074.wav
+20170001P00226I0018 ST-CMDS-20170001_1-OS/20170001P00226I0018.wav
+20170001P00393I0033 ST-CMDS-20170001_1-OS/20170001P00393I0033.wav
+20170001P00335A0009 ST-CMDS-20170001_1-OS/20170001P00335A0009.wav
+20170001P00429A0061 ST-CMDS-20170001_1-OS/20170001P00429A0061.wav
+20170001P00251I0101 ST-CMDS-20170001_1-OS/20170001P00251I0101.wav
+20170001P00214I0060 ST-CMDS-20170001_1-OS/20170001P00214I0060.wav
+20170001P00087A0107 ST-CMDS-20170001_1-OS/20170001P00087A0107.wav
+20170001P00186A0089 ST-CMDS-20170001_1-OS/20170001P00186A0089.wav
+20170001P00112A0039 ST-CMDS-20170001_1-OS/20170001P00112A0039.wav
+20170001P00386A0106 ST-CMDS-20170001_1-OS/20170001P00386A0106.wav
+20170001P00191A0118 ST-CMDS-20170001_1-OS/20170001P00191A0118.wav
+20170001P00089A0117 ST-CMDS-20170001_1-OS/20170001P00089A0117.wav
+20170001P00156A0001 ST-CMDS-20170001_1-OS/20170001P00156A0001.wav
+20170001P00305A0056 ST-CMDS-20170001_1-OS/20170001P00305A0056.wav
+20170001P00088I0021 ST-CMDS-20170001_1-OS/20170001P00088I0021.wav
+20170001P00348I0095 ST-CMDS-20170001_1-OS/20170001P00348I0095.wav
+20170001P00330A0057 ST-CMDS-20170001_1-OS/20170001P00330A0057.wav
+20170001P00423A0007 ST-CMDS-20170001_1-OS/20170001P00423A0007.wav
+20170001P00186I0088 ST-CMDS-20170001_1-OS/20170001P00186I0088.wav
+20170001P00092I0028 ST-CMDS-20170001_1-OS/20170001P00092I0028.wav
+20170001P00126I0115 ST-CMDS-20170001_1-OS/20170001P00126I0115.wav
+20170001P00395I0035 ST-CMDS-20170001_1-OS/20170001P00395I0035.wav
+20170001P00234A0043 ST-CMDS-20170001_1-OS/20170001P00234A0043.wav
+20170001P00347A0079 ST-CMDS-20170001_1-OS/20170001P00347A0079.wav
+20170001P00135I0010 ST-CMDS-20170001_1-OS/20170001P00135I0010.wav
+20170001P00109A0117 ST-CMDS-20170001_1-OS/20170001P00109A0117.wav
+20170001P00263I0015 ST-CMDS-20170001_1-OS/20170001P00263I0015.wav
+20170001P00086I0110 ST-CMDS-20170001_1-OS/20170001P00086I0110.wav
+20170001P00099A0007 ST-CMDS-20170001_1-OS/20170001P00099A0007.wav
+20170001P00066A0067 ST-CMDS-20170001_1-OS/20170001P00066A0067.wav
+20170001P00025I0085 ST-CMDS-20170001_1-OS/20170001P00025I0085.wav
+20170001P00371I0113 ST-CMDS-20170001_1-OS/20170001P00371I0113.wav
+20170001P00330A0035 ST-CMDS-20170001_1-OS/20170001P00330A0035.wav
+20170001P00129I0109 ST-CMDS-20170001_1-OS/20170001P00129I0109.wav
+20170001P00253A0082 ST-CMDS-20170001_1-OS/20170001P00253A0082.wav
+20170001P00063A0064 ST-CMDS-20170001_1-OS/20170001P00063A0064.wav
+20170001P00362A0007 ST-CMDS-20170001_1-OS/20170001P00362A0007.wav
+20170001P00239I0022 ST-CMDS-20170001_1-OS/20170001P00239I0022.wav
+20170001P00203A0031 ST-CMDS-20170001_1-OS/20170001P00203A0031.wav
+20170001P00107A0038 ST-CMDS-20170001_1-OS/20170001P00107A0038.wav
+20170001P00077A0113 ST-CMDS-20170001_1-OS/20170001P00077A0113.wav
+20170001P00307I0082 ST-CMDS-20170001_1-OS/20170001P00307I0082.wav
+20170001P00344A0098 ST-CMDS-20170001_1-OS/20170001P00344A0098.wav
+20170001P00361A0015 ST-CMDS-20170001_1-OS/20170001P00361A0015.wav
+20170001P00120A0038 ST-CMDS-20170001_1-OS/20170001P00120A0038.wav
+20170001P00118A0068 ST-CMDS-20170001_1-OS/20170001P00118A0068.wav
+20170001P00283A0047 ST-CMDS-20170001_1-OS/20170001P00283A0047.wav
+20170001P00135I0066 ST-CMDS-20170001_1-OS/20170001P00135I0066.wav
+20170001P00126A0029 ST-CMDS-20170001_1-OS/20170001P00126A0029.wav
+20170001P00402I0035 ST-CMDS-20170001_1-OS/20170001P00402I0035.wav
+20170001P00171I0037 ST-CMDS-20170001_1-OS/20170001P00171I0037.wav
+20170001P00025A0096 ST-CMDS-20170001_1-OS/20170001P00025A0096.wav
+20170001P00019I0029 ST-CMDS-20170001_1-OS/20170001P00019I0029.wav
+20170001P00261I0051 ST-CMDS-20170001_1-OS/20170001P00261I0051.wav
+20170001P00150I0069 ST-CMDS-20170001_1-OS/20170001P00150I0069.wav
+20170001P00099I0099 ST-CMDS-20170001_1-OS/20170001P00099I0099.wav

datalist/thchs30/cv.wav.lst

@@ -0,0 +1,893 @@
+A11_101 data_thchs30/dev/A11_101.wav
+A11_119 data_thchs30/dev/A11_119.wav
+A11_124 data_thchs30/dev/A11_124.wav
+A11_132 data_thchs30/dev/A11_132.wav
+A11_170 data_thchs30/dev/A11_170.wav
+A11_173 data_thchs30/dev/A11_173.wav
+A11_180 data_thchs30/dev/A11_180.wav
+A11_189 data_thchs30/dev/A11_189.wav
+A11_191 data_thchs30/dev/A11_191.wav
+A11_203 data_thchs30/dev/A11_203.wav
+A11_230 data_thchs30/dev/A11_230.wav
+A11_24 data_thchs30/dev/A11_24.wav
+A11_39 data_thchs30/dev/A11_39.wav
+A11_51 data_thchs30/dev/A11_51.wav
+A11_56 data_thchs30/dev/A11_56.wav
+A11_60 data_thchs30/dev/A11_60.wav
+A11_77 data_thchs30/dev/A11_77.wav
+A11_81 data_thchs30/dev/A11_81.wav
+A12_10 data_thchs30/dev/A12_10.wav
+A12_115 data_thchs30/dev/A12_115.wav
+A12_137 data_thchs30/dev/A12_137.wav
+A12_138 data_thchs30/dev/A12_138.wav
+A12_18 data_thchs30/dev/A12_18.wav
+A12_28 data_thchs30/dev/A12_28.wav
+A12_38 data_thchs30/dev/A12_38.wav
+A12_98 data_thchs30/dev/A12_98.wav
+A13_126 data_thchs30/dev/A13_126.wav
+A13_160 data_thchs30/dev/A13_160.wav
+A13_178 data_thchs30/dev/A13_178.wav
+A13_198 data_thchs30/dev/A13_198.wav
+A13_199 data_thchs30/dev/A13_199.wav
+A13_200 data_thchs30/dev/A13_200.wav
+A13_202 data_thchs30/dev/A13_202.wav
+A13_212 data_thchs30/dev/A13_212.wav
+A13_39 data_thchs30/dev/A13_39.wav
+A13_41 data_thchs30/dev/A13_41.wav
+A13_46 data_thchs30/dev/A13_46.wav
+A13_47 data_thchs30/dev/A13_47.wav
+A13_59 data_thchs30/dev/A13_59.wav
+A13_62 data_thchs30/dev/A13_62.wav
+A13_80 data_thchs30/dev/A13_80.wav
+A13_87 data_thchs30/dev/A13_87.wav
+A13_91 data_thchs30/dev/A13_91.wav
+A13_98 data_thchs30/dev/A13_98.wav
+A14_10 data_thchs30/dev/A14_10.wav
+A14_112 data_thchs30/dev/A14_112.wav
+A14_139 data_thchs30/dev/A14_139.wav
+A14_148 data_thchs30/dev/A14_148.wav
+A14_150 data_thchs30/dev/A14_150.wav
+A14_159 data_thchs30/dev/A14_159.wav
+A14_181 data_thchs30/dev/A14_181.wav
+A14_191 data_thchs30/dev/A14_191.wav
+A14_202 data_thchs30/dev/A14_202.wav
+A14_203 data_thchs30/dev/A14_203.wav
+A14_225 data_thchs30/dev/A14_225.wav
+A14_241 data_thchs30/dev/A14_241.wav
+A14_249 data_thchs30/dev/A14_249.wav
+A14_37 data_thchs30/dev/A14_37.wav
+A14_63 data_thchs30/dev/A14_63.wav
+A14_68 data_thchs30/dev/A14_68.wav
+A14_76 data_thchs30/dev/A14_76.wav
+A14_82 data_thchs30/dev/A14_82.wav
+A19_101 data_thchs30/dev/A19_101.wav
+A19_117 data_thchs30/dev/A19_117.wav
+A19_118 data_thchs30/dev/A19_118.wav
+A19_145 data_thchs30/dev/A19_145.wav
+A19_156 data_thchs30/dev/A19_156.wav
+A19_165 data_thchs30/dev/A19_165.wav
+A19_188 data_thchs30/dev/A19_188.wav
+A19_200 data_thchs30/dev/A19_200.wav
+A19_213 data_thchs30/dev/A19_213.wav
+A19_215 data_thchs30/dev/A19_215.wav
+A19_220 data_thchs30/dev/A19_220.wav
+A19_236 data_thchs30/dev/A19_236.wav
+A19_25 data_thchs30/dev/A19_25.wav
+A19_43 data_thchs30/dev/A19_43.wav
+A19_44 data_thchs30/dev/A19_44.wav
+A19_68 data_thchs30/dev/A19_68.wav
+A19_70 data_thchs30/dev/A19_70.wav
+A19_90 data_thchs30/dev/A19_90.wav
+A22_1 data_thchs30/dev/A22_1.wav
+A22_133 data_thchs30/dev/A22_133.wav
+A22_138 data_thchs30/dev/A22_138.wav
+A22_153 data_thchs30/dev/A22_153.wav
+A22_155 data_thchs30/dev/A22_155.wav
+A22_159 data_thchs30/dev/A22_159.wav
+A22_162 data_thchs30/dev/A22_162.wav
+A22_190 data_thchs30/dev/A22_190.wav
+A22_20 data_thchs30/dev/A22_20.wav
+A22_205 data_thchs30/dev/A22_205.wav
+A22_212 data_thchs30/dev/A22_212.wav
+A22_220 data_thchs30/dev/A22_220.wav
+A22_223 data_thchs30/dev/A22_223.wav
+A22_231 data_thchs30/dev/A22_231.wav
+A22_26 data_thchs30/dev/A22_26.wav
+A22_36 data_thchs30/dev/A22_36.wav
+A22_38 data_thchs30/dev/A22_38.wav
+A22_53 data_thchs30/dev/A22_53.wav
+A22_58 data_thchs30/dev/A22_58.wav
+A22_65 data_thchs30/dev/A22_65.wav
+A22_76 data_thchs30/dev/A22_76.wav
+A22_8 data_thchs30/dev/A22_8.wav
+A22_95 data_thchs30/dev/A22_95.wav
+A23_109 data_thchs30/dev/A23_109.wav
+A23_115 data_thchs30/dev/A23_115.wav
+A23_118 data_thchs30/dev/A23_118.wav
+A23_120 data_thchs30/dev/A23_120.wav
+A23_162 data_thchs30/dev/A23_162.wav
+A23_163 data_thchs30/dev/A23_163.wav
+A23_170 data_thchs30/dev/A23_170.wav
+A23_171 data_thchs30/dev/A23_171.wav
+A23_188 data_thchs30/dev/A23_188.wav
+A23_194 data_thchs30/dev/A23_194.wav
+A23_196 data_thchs30/dev/A23_196.wav
+A23_207 data_thchs30/dev/A23_207.wav
+A23_233 data_thchs30/dev/A23_233.wav
+A23_28 data_thchs30/dev/A23_28.wav
+A23_33 data_thchs30/dev/A23_33.wav
+A23_46 data_thchs30/dev/A23_46.wav
+A23_7 data_thchs30/dev/A23_7.wav
+A23_71 data_thchs30/dev/A23_71.wav
+A23_85 data_thchs30/dev/A23_85.wav
+A23_97 data_thchs30/dev/A23_97.wav
+A2_103 data_thchs30/dev/A2_103.wav
+A2_108 data_thchs30/dev/A2_108.wav
+A2_116 data_thchs30/dev/A2_116.wav
+A2_119 data_thchs30/dev/A2_119.wav
+A2_167 data_thchs30/dev/A2_167.wav
+A2_206 data_thchs30/dev/A2_206.wav
+A2_33 data_thchs30/dev/A2_33.wav
+A2_58 data_thchs30/dev/A2_58.wav
+A2_85 data_thchs30/dev/A2_85.wav
+A2_92 data_thchs30/dev/A2_92.wav
+A32_114 data_thchs30/dev/A32_114.wav
+A32_124 data_thchs30/dev/A32_124.wav
+A32_14 data_thchs30/dev/A32_14.wav
+A32_173 data_thchs30/dev/A32_173.wav
+A32_19 data_thchs30/dev/A32_19.wav
+A32_202 data_thchs30/dev/A32_202.wav
+A32_219 data_thchs30/dev/A32_219.wav
+A32_232 data_thchs30/dev/A32_232.wav
+A32_235 data_thchs30/dev/A32_235.wav
+A32_244 data_thchs30/dev/A32_244.wav
+A32_35 data_thchs30/dev/A32_35.wav
+A32_37 data_thchs30/dev/A32_37.wav
+A32_50 data_thchs30/dev/A32_50.wav
+A32_54 data_thchs30/dev/A32_54.wav
+A32_62 data_thchs30/dev/A32_62.wav
+A32_67 data_thchs30/dev/A32_67.wav
+A32_84 data_thchs30/dev/A32_84.wav
+A32_91 data_thchs30/dev/A32_91.wav
+A33_11 data_thchs30/dev/A33_11.wav
+A33_20 data_thchs30/dev/A33_20.wav
+A33_3 data_thchs30/dev/A33_3.wav
+A33_36 data_thchs30/dev/A33_36.wav
+A33_79 data_thchs30/dev/A33_79.wav
+A34_118 data_thchs30/dev/A34_118.wav
+A34_120 data_thchs30/dev/A34_120.wav
+A34_127 data_thchs30/dev/A34_127.wav
+A34_132 data_thchs30/dev/A34_132.wav
+A34_134 data_thchs30/dev/A34_134.wav
+A34_160 data_thchs30/dev/A34_160.wav
+A34_176 data_thchs30/dev/A34_176.wav
+A34_185 data_thchs30/dev/A34_185.wav
+A34_190 data_thchs30/dev/A34_190.wav
+A34_194 data_thchs30/dev/A34_194.wav
+A34_20 data_thchs30/dev/A34_20.wav
+A34_203 data_thchs30/dev/A34_203.wav
+A34_237 data_thchs30/dev/A34_237.wav
+A34_244 data_thchs30/dev/A34_244.wav
+A34_29 data_thchs30/dev/A34_29.wav
+A34_36 data_thchs30/dev/A34_36.wav
+A34_51 data_thchs30/dev/A34_51.wav
+A34_63 data_thchs30/dev/A34_63.wav
+A34_73 data_thchs30/dev/A34_73.wav
+A34_75 data_thchs30/dev/A34_75.wav
+A35_2 data_thchs30/dev/A35_2.wav
+A36_100 data_thchs30/dev/A36_100.wav
+A36_115 data_thchs30/dev/A36_115.wav
+A36_133 data_thchs30/dev/A36_133.wav
+A36_151 data_thchs30/dev/A36_151.wav
+A36_158 data_thchs30/dev/A36_158.wav
+A36_163 data_thchs30/dev/A36_163.wav
+A36_177 data_thchs30/dev/A36_177.wav
+A36_203 data_thchs30/dev/A36_203.wav
+A36_208 data_thchs30/dev/A36_208.wav
+A36_233 data_thchs30/dev/A36_233.wav
+A36_245 data_thchs30/dev/A36_245.wav
+A36_40 data_thchs30/dev/A36_40.wav
+A36_52 data_thchs30/dev/A36_52.wav
+A36_59 data_thchs30/dev/A36_59.wav
+A36_68 data_thchs30/dev/A36_68.wav
+A36_88 data_thchs30/dev/A36_88.wav
+A4_10 data_thchs30/dev/A4_10.wav
+A4_125 data_thchs30/dev/A4_125.wav
+A4_140 data_thchs30/dev/A4_140.wav
+A4_157 data_thchs30/dev/A4_157.wav
+A4_182 data_thchs30/dev/A4_182.wav
+A4_197 data_thchs30/dev/A4_197.wav
+A4_204 data_thchs30/dev/A4_204.wav
+A4_209 data_thchs30/dev/A4_209.wav
+A4_210 data_thchs30/dev/A4_210.wav
+A4_235 data_thchs30/dev/A4_235.wav
+A4_240 data_thchs30/dev/A4_240.wav
+A4_243 data_thchs30/dev/A4_243.wav
+A4_26 data_thchs30/dev/A4_26.wav
+A4_4 data_thchs30/dev/A4_4.wav
+A4_48 data_thchs30/dev/A4_48.wav
+A4_52 data_thchs30/dev/A4_52.wav
+A4_74 data_thchs30/dev/A4_74.wav
+A6_0 data_thchs30/dev/A6_0.wav
+A6_108 data_thchs30/dev/A6_108.wav
+A6_109 data_thchs30/dev/A6_109.wav
+A6_11 data_thchs30/dev/A6_11.wav
+A6_128 data_thchs30/dev/A6_128.wav
+A6_138 data_thchs30/dev/A6_138.wav
+A6_141 data_thchs30/dev/A6_141.wav
+A6_145 data_thchs30/dev/A6_145.wav
+A6_16 data_thchs30/dev/A6_16.wav
+A6_171 data_thchs30/dev/A6_171.wav
+A6_20 data_thchs30/dev/A6_20.wav
+A6_205 data_thchs30/dev/A6_205.wav
+A6_212 data_thchs30/dev/A6_212.wav
+A6_231 data_thchs30/dev/A6_231.wav
+A6_247 data_thchs30/dev/A6_247.wav
+A6_3 data_thchs30/dev/A6_3.wav
+A6_41 data_thchs30/dev/A6_41.wav
+A6_77 data_thchs30/dev/A6_77.wav
+A6_78 data_thchs30/dev/A6_78.wav
+A6_89 data_thchs30/dev/A6_89.wav
+A7_1 data_thchs30/dev/A7_1.wav
+A7_109 data_thchs30/dev/A7_109.wav
+A7_132 data_thchs30/dev/A7_132.wav
+A7_135 data_thchs30/dev/A7_135.wav
+A7_157 data_thchs30/dev/A7_157.wav
+A7_162 data_thchs30/dev/A7_162.wav
+A7_185 data_thchs30/dev/A7_185.wav
+A7_194 data_thchs30/dev/A7_194.wav
+A7_197 data_thchs30/dev/A7_197.wav
+A7_199 data_thchs30/dev/A7_199.wav
+A7_205 data_thchs30/dev/A7_205.wav
+A7_211 data_thchs30/dev/A7_211.wav
+A7_218 data_thchs30/dev/A7_218.wav
+A7_219 data_thchs30/dev/A7_219.wav
+A7_229 data_thchs30/dev/A7_229.wav
+A7_29 data_thchs30/dev/A7_29.wav
+A7_40 data_thchs30/dev/A7_40.wav
+A7_62 data_thchs30/dev/A7_62.wav
+A7_93 data_thchs30/dev/A7_93.wav
+A8_1 data_thchs30/dev/A8_1.wav
+A8_107 data_thchs30/dev/A8_107.wav
+A8_108 data_thchs30/dev/A8_108.wav
+A8_115 data_thchs30/dev/A8_115.wav
+A8_142 data_thchs30/dev/A8_142.wav
+A8_152 data_thchs30/dev/A8_152.wav
+A8_154 data_thchs30/dev/A8_154.wav
+A8_171 data_thchs30/dev/A8_171.wav
+A8_173 data_thchs30/dev/A8_173.wav
+A8_181 data_thchs30/dev/A8_181.wav
+A8_201 data_thchs30/dev/A8_201.wav
+A8_209 data_thchs30/dev/A8_209.wav
+A8_235 data_thchs30/dev/A8_235.wav
+A8_33 data_thchs30/dev/A8_33.wav
+A8_40 data_thchs30/dev/A8_40.wav
+A8_51 data_thchs30/dev/A8_51.wav
+A8_67 data_thchs30/dev/A8_67.wav
+A8_77 data_thchs30/dev/A8_77.wav
+A8_8 data_thchs30/dev/A8_8.wav
+A8_92 data_thchs30/dev/A8_92.wav
+A8_98 data_thchs30/dev/A8_98.wav
+A9_0 data_thchs30/dev/A9_0.wav
+B11_251 data_thchs30/dev/B11_251.wav
+B11_253 data_thchs30/dev/B11_253.wav
+B11_255 data_thchs30/dev/B11_255.wav
+B11_268 data_thchs30/dev/B11_268.wav
+B11_278 data_thchs30/dev/B11_278.wav
+B11_330 data_thchs30/dev/B11_330.wav
+B11_368 data_thchs30/dev/B11_368.wav
+B11_397 data_thchs30/dev/B11_397.wav
+B11_417 data_thchs30/dev/B11_417.wav
+B11_432 data_thchs30/dev/B11_432.wav
+B11_435 data_thchs30/dev/B11_435.wav
+B11_440 data_thchs30/dev/B11_440.wav
+B11_447 data_thchs30/dev/B11_447.wav
+B11_460 data_thchs30/dev/B11_460.wav
+B11_462 data_thchs30/dev/B11_462.wav
+B11_475 data_thchs30/dev/B11_475.wav
+B11_495 data_thchs30/dev/B11_495.wav
+B12_256 data_thchs30/dev/B12_256.wav
+B12_270 data_thchs30/dev/B12_270.wav
+B12_285 data_thchs30/dev/B12_285.wav
+B12_305 data_thchs30/dev/B12_305.wav
+B12_344 data_thchs30/dev/B12_344.wav
+B12_350 data_thchs30/dev/B12_350.wav
+B12_371 data_thchs30/dev/B12_371.wav
+B12_378 data_thchs30/dev/B12_378.wav
+B12_380 data_thchs30/dev/B12_380.wav
+B12_384 data_thchs30/dev/B12_384.wav
+B12_387 data_thchs30/dev/B12_387.wav
+B12_396 data_thchs30/dev/B12_396.wav
+B12_397 data_thchs30/dev/B12_397.wav
+B12_400 data_thchs30/dev/B12_400.wav
+B12_438 data_thchs30/dev/B12_438.wav
+B12_448 data_thchs30/dev/B12_448.wav
+B12_459 data_thchs30/dev/B12_459.wav
+B12_461 data_thchs30/dev/B12_461.wav
+B12_486 data_thchs30/dev/B12_486.wav
+B12_489 data_thchs30/dev/B12_489.wav
+B15_255 data_thchs30/dev/B15_255.wav
+B15_265 data_thchs30/dev/B15_265.wav
+B15_266 data_thchs30/dev/B15_266.wav
+B15_276 data_thchs30/dev/B15_276.wav
+B15_305 data_thchs30/dev/B15_305.wav
+B15_310 data_thchs30/dev/B15_310.wav
+B15_324 data_thchs30/dev/B15_324.wav
+B15_329 data_thchs30/dev/B15_329.wav
+B15_330 data_thchs30/dev/B15_330.wav
+B15_346 data_thchs30/dev/B15_346.wav
+B15_353 data_thchs30/dev/B15_353.wav
+B15_359 data_thchs30/dev/B15_359.wav
+B15_366 data_thchs30/dev/B15_366.wav
+B15_369 data_thchs30/dev/B15_369.wav
+B15_372 data_thchs30/dev/B15_372.wav
+B15_374 data_thchs30/dev/B15_374.wav
+B15_386 data_thchs30/dev/B15_386.wav
+B15_387 data_thchs30/dev/B15_387.wav
+B15_388 data_thchs30/dev/B15_388.wav
+B15_390 data_thchs30/dev/B15_390.wav
+B15_412 data_thchs30/dev/B15_412.wav
+B15_428 data_thchs30/dev/B15_428.wav
+B15_439 data_thchs30/dev/B15_439.wav
+B15_470 data_thchs30/dev/B15_470.wav
+B15_475 data_thchs30/dev/B15_475.wav
+B15_476 data_thchs30/dev/B15_476.wav
+B15_487 data_thchs30/dev/B15_487.wav
+B22_255 data_thchs30/dev/B22_255.wav
+B22_256 data_thchs30/dev/B22_256.wav
+B22_260 data_thchs30/dev/B22_260.wav
+B22_265 data_thchs30/dev/B22_265.wav
+B22_272 data_thchs30/dev/B22_272.wav
+B22_282 data_thchs30/dev/B22_282.wav
+B22_285 data_thchs30/dev/B22_285.wav
+B22_293 data_thchs30/dev/B22_293.wav
+B22_297 data_thchs30/dev/B22_297.wav
+B22_318 data_thchs30/dev/B22_318.wav
+B22_325 data_thchs30/dev/B22_325.wav
+B22_326 data_thchs30/dev/B22_326.wav
+B22_333 data_thchs30/dev/B22_333.wav
+B22_339 data_thchs30/dev/B22_339.wav
+B22_340 data_thchs30/dev/B22_340.wav
+B22_350 data_thchs30/dev/B22_350.wav
+B22_352 data_thchs30/dev/B22_352.wav
+B22_355 data_thchs30/dev/B22_355.wav
+B22_367 data_thchs30/dev/B22_367.wav
+B22_370 data_thchs30/dev/B22_370.wav
+B22_422 data_thchs30/dev/B22_422.wav
+B22_424 data_thchs30/dev/B22_424.wav
+B22_430 data_thchs30/dev/B22_430.wav
+B22_437 data_thchs30/dev/B22_437.wav
+B22_455 data_thchs30/dev/B22_455.wav
+B22_457 data_thchs30/dev/B22_457.wav
+B22_460 data_thchs30/dev/B22_460.wav
+B22_482 data_thchs30/dev/B22_482.wav
+B22_487 data_thchs30/dev/B22_487.wav
+B22_493 data_thchs30/dev/B22_493.wav
+B22_495 data_thchs30/dev/B22_495.wav
+B2_268 data_thchs30/dev/B2_268.wav
+B2_275 data_thchs30/dev/B2_275.wav
+B2_282 data_thchs30/dev/B2_282.wav
+B2_286 data_thchs30/dev/B2_286.wav
+B2_297 data_thchs30/dev/B2_297.wav
+B2_302 data_thchs30/dev/B2_302.wav
+B2_307 data_thchs30/dev/B2_307.wav
+B2_313 data_thchs30/dev/B2_313.wav
+B2_319 data_thchs30/dev/B2_319.wav
+B2_320 data_thchs30/dev/B2_320.wav
+B2_332 data_thchs30/dev/B2_332.wav
+B2_335 data_thchs30/dev/B2_335.wav
+B2_337 data_thchs30/dev/B2_337.wav
+B2_346 data_thchs30/dev/B2_346.wav
+B2_348 data_thchs30/dev/B2_348.wav
+B2_352 data_thchs30/dev/B2_352.wav
+B2_381 data_thchs30/dev/B2_381.wav
+B2_382 data_thchs30/dev/B2_382.wav
+B2_404 data_thchs30/dev/B2_404.wav
+B2_416 data_thchs30/dev/B2_416.wav
+B2_435 data_thchs30/dev/B2_435.wav
+B2_436 data_thchs30/dev/B2_436.wav
+B2_437 data_thchs30/dev/B2_437.wav
+B2_461 data_thchs30/dev/B2_461.wav
+B2_469 data_thchs30/dev/B2_469.wav
+B2_472 data_thchs30/dev/B2_472.wav
+B2_477 data_thchs30/dev/B2_477.wav
+B2_488 data_thchs30/dev/B2_488.wav
+B2_492 data_thchs30/dev/B2_492.wav
+B31_260 data_thchs30/dev/B31_260.wav
+B31_279 data_thchs30/dev/B31_279.wav
+B31_291 data_thchs30/dev/B31_291.wav
+B31_297 data_thchs30/dev/B31_297.wav
+B31_320 data_thchs30/dev/B31_320.wav
+B31_343 data_thchs30/dev/B31_343.wav
+B31_377 data_thchs30/dev/B31_377.wav
+B31_384 data_thchs30/dev/B31_384.wav
+B31_385 data_thchs30/dev/B31_385.wav
+B31_388 data_thchs30/dev/B31_388.wav
+B31_389 data_thchs30/dev/B31_389.wav
+B31_403 data_thchs30/dev/B31_403.wav
+B31_413 data_thchs30/dev/B31_413.wav
+B31_416 data_thchs30/dev/B31_416.wav
+B31_418 data_thchs30/dev/B31_418.wav
+B31_436 data_thchs30/dev/B31_436.wav
+B31_439 data_thchs30/dev/B31_439.wav
+B31_455 data_thchs30/dev/B31_455.wav
+B31_460 data_thchs30/dev/B31_460.wav
+B31_474 data_thchs30/dev/B31_474.wav
+B31_476 data_thchs30/dev/B31_476.wav
+B31_481 data_thchs30/dev/B31_481.wav
+B32_275 data_thchs30/dev/B32_275.wav
+B32_294 data_thchs30/dev/B32_294.wav
+B32_296 data_thchs30/dev/B32_296.wav
+B32_297 data_thchs30/dev/B32_297.wav
+B32_301 data_thchs30/dev/B32_301.wav
+B32_302 data_thchs30/dev/B32_302.wav
+B32_303 data_thchs30/dev/B32_303.wav
+B32_304 data_thchs30/dev/B32_304.wav
+B32_321 data_thchs30/dev/B32_321.wav
+B32_323 data_thchs30/dev/B32_323.wav
+B32_329 data_thchs30/dev/B32_329.wav
+B32_335 data_thchs30/dev/B32_335.wav
+B32_342 data_thchs30/dev/B32_342.wav
+B32_356 data_thchs30/dev/B32_356.wav
+B32_364 data_thchs30/dev/B32_364.wav
+B32_365 data_thchs30/dev/B32_365.wav
+B32_369 data_thchs30/dev/B32_369.wav
+B32_412 data_thchs30/dev/B32_412.wav
+B32_427 data_thchs30/dev/B32_427.wav
+B32_429 data_thchs30/dev/B32_429.wav
+B32_491 data_thchs30/dev/B32_491.wav
+B33_250 data_thchs30/dev/B33_250.wav
+B33_253 data_thchs30/dev/B33_253.wav
+B33_269 data_thchs30/dev/B33_269.wav
+B33_272 data_thchs30/dev/B33_272.wav
+B33_278 data_thchs30/dev/B33_278.wav
+B33_298 data_thchs30/dev/B33_298.wav
+B33_352 data_thchs30/dev/B33_352.wav
+B33_355 data_thchs30/dev/B33_355.wav
+B33_363 data_thchs30/dev/B33_363.wav
+B33_389 data_thchs30/dev/B33_389.wav
+B33_390 data_thchs30/dev/B33_390.wav
+B33_396 data_thchs30/dev/B33_396.wav
+B33_400 data_thchs30/dev/B33_400.wav
+B33_406 data_thchs30/dev/B33_406.wav
+B33_409 data_thchs30/dev/B33_409.wav
+B33_428 data_thchs30/dev/B33_428.wav
+B33_441 data_thchs30/dev/B33_441.wav
+B33_458 data_thchs30/dev/B33_458.wav
+B33_470 data_thchs30/dev/B33_470.wav
+B33_474 data_thchs30/dev/B33_474.wav
+B34_251 data_thchs30/dev/B34_251.wav
+B34_254 data_thchs30/dev/B34_254.wav
+B4_266 data_thchs30/dev/B4_266.wav
+B4_275 data_thchs30/dev/B4_275.wav
+B4_319 data_thchs30/dev/B4_319.wav
+B4_320 data_thchs30/dev/B4_320.wav
+B4_352 data_thchs30/dev/B4_352.wav
+B4_354 data_thchs30/dev/B4_354.wav
+B4_357 data_thchs30/dev/B4_357.wav
+B4_364 data_thchs30/dev/B4_364.wav
+B4_398 data_thchs30/dev/B4_398.wav
+B4_399 data_thchs30/dev/B4_399.wav
+B4_417 data_thchs30/dev/B4_417.wav
+B4_418 data_thchs30/dev/B4_418.wav
+B4_423 data_thchs30/dev/B4_423.wav
+B4_431 data_thchs30/dev/B4_431.wav
+B4_434 data_thchs30/dev/B4_434.wav
+B4_460 data_thchs30/dev/B4_460.wav
+B4_464 data_thchs30/dev/B4_464.wav
+B4_475 data_thchs30/dev/B4_475.wav
+B4_491 data_thchs30/dev/B4_491.wav
+B6_251 data_thchs30/dev/B6_251.wav
+B6_310 data_thchs30/dev/B6_310.wav
+B6_319 data_thchs30/dev/B6_319.wav
+B6_323 data_thchs30/dev/B6_323.wav
+B6_328 data_thchs30/dev/B6_328.wav
+B6_335 data_thchs30/dev/B6_335.wav
+B6_356 data_thchs30/dev/B6_356.wav
+B6_398 data_thchs30/dev/B6_398.wav
+B6_420 data_thchs30/dev/B6_420.wav
+B6_423 data_thchs30/dev/B6_423.wav
+B6_449 data_thchs30/dev/B6_449.wav
+B6_455 data_thchs30/dev/B6_455.wav
+B6_462 data_thchs30/dev/B6_462.wav
+B6_468 data_thchs30/dev/B6_468.wav
+B6_471 data_thchs30/dev/B6_471.wav
+B6_476 data_thchs30/dev/B6_476.wav
+B6_482 data_thchs30/dev/B6_482.wav
+B6_493 data_thchs30/dev/B6_493.wav
+B7_251 data_thchs30/dev/B7_251.wav
+B7_261 data_thchs30/dev/B7_261.wav
+B7_278 data_thchs30/dev/B7_278.wav
+B7_279 data_thchs30/dev/B7_279.wav
+B7_289 data_thchs30/dev/B7_289.wav
+B7_313 data_thchs30/dev/B7_313.wav
+B7_315 data_thchs30/dev/B7_315.wav
+B7_351 data_thchs30/dev/B7_351.wav
+B7_358 data_thchs30/dev/B7_358.wav
+B7_368 data_thchs30/dev/B7_368.wav
+B7_369 data_thchs30/dev/B7_369.wav
+B7_384 data_thchs30/dev/B7_384.wav
+B7_389 data_thchs30/dev/B7_389.wav
+B7_395 data_thchs30/dev/B7_395.wav
+B7_420 data_thchs30/dev/B7_420.wav
+B7_421 data_thchs30/dev/B7_421.wav
+B7_437 data_thchs30/dev/B7_437.wav
+B7_440 data_thchs30/dev/B7_440.wav
+B7_441 data_thchs30/dev/B7_441.wav
+B7_451 data_thchs30/dev/B7_451.wav
+B7_456 data_thchs30/dev/B7_456.wav
+B7_463 data_thchs30/dev/B7_463.wav
+B7_467 data_thchs30/dev/B7_467.wav
+B7_468 data_thchs30/dev/B7_468.wav
+B8_256 data_thchs30/dev/B8_256.wav
+B8_272 data_thchs30/dev/B8_272.wav
+B8_285 data_thchs30/dev/B8_285.wav
+B8_287 data_thchs30/dev/B8_287.wav
+B8_295 data_thchs30/dev/B8_295.wav
+B8_307 data_thchs30/dev/B8_307.wav
+B8_330 data_thchs30/dev/B8_330.wav
+B8_359 data_thchs30/dev/B8_359.wav
+B8_375 data_thchs30/dev/B8_375.wav
+B8_384 data_thchs30/dev/B8_384.wav
+B8_408 data_thchs30/dev/B8_408.wav
+B8_415 data_thchs30/dev/B8_415.wav
+B8_419 data_thchs30/dev/B8_419.wav
+B8_423 data_thchs30/dev/B8_423.wav
+B8_455 data_thchs30/dev/B8_455.wav
+B8_467 data_thchs30/dev/B8_467.wav
+B8_472 data_thchs30/dev/B8_472.wav
+B8_478 data_thchs30/dev/B8_478.wav
+B8_480 data_thchs30/dev/B8_480.wav
+B8_483 data_thchs30/dev/B8_483.wav
+B8_498 data_thchs30/dev/B8_498.wav
+C12_511 data_thchs30/dev/C12_511.wav
+C12_518 data_thchs30/dev/C12_518.wav
+C12_526 data_thchs30/dev/C12_526.wav
+C12_544 data_thchs30/dev/C12_544.wav
+C12_549 data_thchs30/dev/C12_549.wav
+C12_595 data_thchs30/dev/C12_595.wav
+C12_603 data_thchs30/dev/C12_603.wav
+C12_615 data_thchs30/dev/C12_615.wav
+C12_616 data_thchs30/dev/C12_616.wav
+C12_628 data_thchs30/dev/C12_628.wav
+C12_637 data_thchs30/dev/C12_637.wav
+C12_670 data_thchs30/dev/C12_670.wav
+C12_680 data_thchs30/dev/C12_680.wav
+C12_697 data_thchs30/dev/C12_697.wav
+C13_500 data_thchs30/dev/C13_500.wav
+C13_505 data_thchs30/dev/C13_505.wav
+C13_519 data_thchs30/dev/C13_519.wav
+C13_521 data_thchs30/dev/C13_521.wav
+C13_539 data_thchs30/dev/C13_539.wav
+C13_560 data_thchs30/dev/C13_560.wav
+C13_573 data_thchs30/dev/C13_573.wav
+C13_577 data_thchs30/dev/C13_577.wav
+C13_587 data_thchs30/dev/C13_587.wav
+C13_589 data_thchs30/dev/C13_589.wav
+C13_611 data_thchs30/dev/C13_611.wav
+C13_634 data_thchs30/dev/C13_634.wav
+C13_687 data_thchs30/dev/C13_687.wav
+C13_694 data_thchs30/dev/C13_694.wav
+C13_706 data_thchs30/dev/C13_706.wav
+C13_722 data_thchs30/dev/C13_722.wav
+C14_501 data_thchs30/dev/C14_501.wav
+C14_516 data_thchs30/dev/C14_516.wav
+C14_518 data_thchs30/dev/C14_518.wav
+C14_524 data_thchs30/dev/C14_524.wav
+C14_529 data_thchs30/dev/C14_529.wav
+C14_537 data_thchs30/dev/C14_537.wav
+C14_556 data_thchs30/dev/C14_556.wav
+C14_562 data_thchs30/dev/C14_562.wav
+C14_585 data_thchs30/dev/C14_585.wav
+C14_594 data_thchs30/dev/C14_594.wav
+C14_596 data_thchs30/dev/C14_596.wav
+C14_600 data_thchs30/dev/C14_600.wav
+C14_627 data_thchs30/dev/C14_627.wav
+C14_628 data_thchs30/dev/C14_628.wav
+C14_630 data_thchs30/dev/C14_630.wav
+C14_635 data_thchs30/dev/C14_635.wav
+C14_638 data_thchs30/dev/C14_638.wav
+C14_647 data_thchs30/dev/C14_647.wav
+C14_664 data_thchs30/dev/C14_664.wav
+C14_669 data_thchs30/dev/C14_669.wav
+C14_670 data_thchs30/dev/C14_670.wav
+C14_677 data_thchs30/dev/C14_677.wav
+C14_713 data_thchs30/dev/C14_713.wav
+C14_724 data_thchs30/dev/C14_724.wav
+C14_725 data_thchs30/dev/C14_725.wav
+C14_747 data_thchs30/dev/C14_747.wav
+C17_529 data_thchs30/dev/C17_529.wav
+C17_556 data_thchs30/dev/C17_556.wav
+C17_567 data_thchs30/dev/C17_567.wav
+C17_592 data_thchs30/dev/C17_592.wav
+C17_596 data_thchs30/dev/C17_596.wav
+C17_597 data_thchs30/dev/C17_597.wav
+C17_600 data_thchs30/dev/C17_600.wav
+C17_603 data_thchs30/dev/C17_603.wav
+C17_618 data_thchs30/dev/C17_618.wav
+C17_627 data_thchs30/dev/C17_627.wav
+C17_641 data_thchs30/dev/C17_641.wav
+C17_669 data_thchs30/dev/C17_669.wav
+C17_671 data_thchs30/dev/C17_671.wav
+C17_678 data_thchs30/dev/C17_678.wav
+C17_689 data_thchs30/dev/C17_689.wav
+C17_695 data_thchs30/dev/C17_695.wav
+C17_736 data_thchs30/dev/C17_736.wav
+C17_748 data_thchs30/dev/C17_748.wav
+C18_502 data_thchs30/dev/C18_502.wav
+C18_509 data_thchs30/dev/C18_509.wav
+C18_538 data_thchs30/dev/C18_538.wav
+C18_546 data_thchs30/dev/C18_546.wav
+C18_548 data_thchs30/dev/C18_548.wav
+C18_553 data_thchs30/dev/C18_553.wav
+C18_556 data_thchs30/dev/C18_556.wav
+C18_557 data_thchs30/dev/C18_557.wav
+C18_573 data_thchs30/dev/C18_573.wav
+C18_588 data_thchs30/dev/C18_588.wav
+C18_593 data_thchs30/dev/C18_593.wav
+C18_594 data_thchs30/dev/C18_594.wav
+C18_599 data_thchs30/dev/C18_599.wav
+C18_606 data_thchs30/dev/C18_606.wav
+C18_619 data_thchs30/dev/C18_619.wav
+C18_626 data_thchs30/dev/C18_626.wav
+C18_651 data_thchs30/dev/C18_651.wav
+C18_656 data_thchs30/dev/C18_656.wav
+C18_664 data_thchs30/dev/C18_664.wav
+C18_677 data_thchs30/dev/C18_677.wav
+C18_695 data_thchs30/dev/C18_695.wav
+C18_737 data_thchs30/dev/C18_737.wav
+C18_745 data_thchs30/dev/C18_745.wav
+C18_748 data_thchs30/dev/C18_748.wav
+C19_509 data_thchs30/dev/C19_509.wav
+C19_530 data_thchs30/dev/C19_530.wav
+C19_536 data_thchs30/dev/C19_536.wav
+C19_567 data_thchs30/dev/C19_567.wav
+C19_570 data_thchs30/dev/C19_570.wav
+C19_593 data_thchs30/dev/C19_593.wav
+C19_607 data_thchs30/dev/C19_607.wav
+C19_613 data_thchs30/dev/C19_613.wav
+C19_617 data_thchs30/dev/C19_617.wav
+C19_622 data_thchs30/dev/C19_622.wav
+C19_639 data_thchs30/dev/C19_639.wav
+C19_665 data_thchs30/dev/C19_665.wav
+C19_678 data_thchs30/dev/C19_678.wav
+C19_684 data_thchs30/dev/C19_684.wav
+C19_712 data_thchs30/dev/C19_712.wav
+C19_717 data_thchs30/dev/C19_717.wav
+C19_723 data_thchs30/dev/C19_723.wav
+C20_511 data_thchs30/dev/C20_511.wav
+C20_515 data_thchs30/dev/C20_515.wav
+C20_521 data_thchs30/dev/C20_521.wav
+C20_530 data_thchs30/dev/C20_530.wav
+C20_531 data_thchs30/dev/C20_531.wav
+C20_561 data_thchs30/dev/C20_561.wav
+C20_567 data_thchs30/dev/C20_567.wav
+C20_587 data_thchs30/dev/C20_587.wav
+C20_588 data_thchs30/dev/C20_588.wav
+C20_606 data_thchs30/dev/C20_606.wav
+C20_610 data_thchs30/dev/C20_610.wav
+C20_612 data_thchs30/dev/C20_612.wav
+C20_613 data_thchs30/dev/C20_613.wav
+C20_616 data_thchs30/dev/C20_616.wav
+C20_624 data_thchs30/dev/C20_624.wav
+C20_626 data_thchs30/dev/C20_626.wav
+C20_628 data_thchs30/dev/C20_628.wav
+C20_635 data_thchs30/dev/C20_635.wav
+C20_638 data_thchs30/dev/C20_638.wav
+C20_641 data_thchs30/dev/C20_641.wav
+C20_643 data_thchs30/dev/C20_643.wav
+C20_657 data_thchs30/dev/C20_657.wav
+C20_658 data_thchs30/dev/C20_658.wav
+C20_686 data_thchs30/dev/C20_686.wav
+C20_699 data_thchs30/dev/C20_699.wav
+C20_729 data_thchs30/dev/C20_729.wav
+C20_747 data_thchs30/dev/C20_747.wav
+C21_500 data_thchs30/dev/C21_500.wav
+C21_505 data_thchs30/dev/C21_505.wav
+C21_520 data_thchs30/dev/C21_520.wav
+C21_523 data_thchs30/dev/C21_523.wav
+C21_526 data_thchs30/dev/C21_526.wav
+C21_530 data_thchs30/dev/C21_530.wav
+C21_571 data_thchs30/dev/C21_571.wav
+C21_584 data_thchs30/dev/C21_584.wav
+C21_603 data_thchs30/dev/C21_603.wav
+C21_604 data_thchs30/dev/C21_604.wav
+C21_606 data_thchs30/dev/C21_606.wav
+C21_613 data_thchs30/dev/C21_613.wav
+C21_632 data_thchs30/dev/C21_632.wav
+C21_638 data_thchs30/dev/C21_638.wav
+C21_640 data_thchs30/dev/C21_640.wav
+C21_653 data_thchs30/dev/C21_653.wav
+C21_670 data_thchs30/dev/C21_670.wav
+C21_684 data_thchs30/dev/C21_684.wav
+C21_692 data_thchs30/dev/C21_692.wav
+C21_709 data_thchs30/dev/C21_709.wav
+C21_721 data_thchs30/dev/C21_721.wav
+C21_730 data_thchs30/dev/C21_730.wav
+C22_509 data_thchs30/dev/C22_509.wav
+C22_512 data_thchs30/dev/C22_512.wav
+C22_516 data_thchs30/dev/C22_516.wav
+C22_519 data_thchs30/dev/C22_519.wav
+C22_542 data_thchs30/dev/C22_542.wav
+C22_563 data_thchs30/dev/C22_563.wav
+C22_578 data_thchs30/dev/C22_578.wav
+C22_579 data_thchs30/dev/C22_579.wav
+C22_594 data_thchs30/dev/C22_594.wav
+C22_596 data_thchs30/dev/C22_596.wav
+C22_603 data_thchs30/dev/C22_603.wav
+C22_604 data_thchs30/dev/C22_604.wav
+C22_610 data_thchs30/dev/C22_610.wav
+C22_622 data_thchs30/dev/C22_622.wav
+C22_629 data_thchs30/dev/C22_629.wav
+C22_630 data_thchs30/dev/C22_630.wav
+C22_635 data_thchs30/dev/C22_635.wav
+C22_649 data_thchs30/dev/C22_649.wav
+C22_650 data_thchs30/dev/C22_650.wav
+C22_656 data_thchs30/dev/C22_656.wav
+C22_658 data_thchs30/dev/C22_658.wav
+C22_673 data_thchs30/dev/C22_673.wav
+C22_675 data_thchs30/dev/C22_675.wav
+C22_682 data_thchs30/dev/C22_682.wav
+C22_685 data_thchs30/dev/C22_685.wav
+C22_693 data_thchs30/dev/C22_693.wav
+C22_697 data_thchs30/dev/C22_697.wav
+C22_702 data_thchs30/dev/C22_702.wav
+C22_734 data_thchs30/dev/C22_734.wav
+C23_512 data_thchs30/dev/C23_512.wav
+C23_532 data_thchs30/dev/C23_532.wav
+C23_549 data_thchs30/dev/C23_549.wav
+C23_553 data_thchs30/dev/C23_553.wav
+C23_558 data_thchs30/dev/C23_558.wav
+C23_571 data_thchs30/dev/C23_571.wav
+C23_572 data_thchs30/dev/C23_572.wav
+C23_575 data_thchs30/dev/C23_575.wav
+C23_592 data_thchs30/dev/C23_592.wav
+C23_597 data_thchs30/dev/C23_597.wav
+C23_598 data_thchs30/dev/C23_598.wav
+C23_606 data_thchs30/dev/C23_606.wav
+C23_612 data_thchs30/dev/C23_612.wav
+C23_622 data_thchs30/dev/C23_622.wav
+C23_628 data_thchs30/dev/C23_628.wav
+C23_636 data_thchs30/dev/C23_636.wav
+C23_639 data_thchs30/dev/C23_639.wav
+C23_654 data_thchs30/dev/C23_654.wav
+C23_657 data_thchs30/dev/C23_657.wav
+C23_667 data_thchs30/dev/C23_667.wav
+C23_681 data_thchs30/dev/C23_681.wav
+C23_686 data_thchs30/dev/C23_686.wav
+C23_700 data_thchs30/dev/C23_700.wav
+C23_708 data_thchs30/dev/C23_708.wav
+C23_711 data_thchs30/dev/C23_711.wav
+C23_715 data_thchs30/dev/C23_715.wav
+C23_720 data_thchs30/dev/C23_720.wav
+C23_724 data_thchs30/dev/C23_724.wav
+C23_729 data_thchs30/dev/C23_729.wav
+C23_738 data_thchs30/dev/C23_738.wav
+C2_506 data_thchs30/dev/C2_506.wav
+C2_511 data_thchs30/dev/C2_511.wav
+C2_519 data_thchs30/dev/C2_519.wav
+C2_534 data_thchs30/dev/C2_534.wav
+C2_542 data_thchs30/dev/C2_542.wav
+C2_556 data_thchs30/dev/C2_556.wav
+C2_565 data_thchs30/dev/C2_565.wav
+C2_572 data_thchs30/dev/C2_572.wav
+C31_506 data_thchs30/dev/C31_506.wav
+C31_508 data_thchs30/dev/C31_508.wav
+C31_528 data_thchs30/dev/C31_528.wav
+C31_538 data_thchs30/dev/C31_538.wav
+C31_540 data_thchs30/dev/C31_540.wav
+C31_574 data_thchs30/dev/C31_574.wav
+C31_578 data_thchs30/dev/C31_578.wav
+C31_590 data_thchs30/dev/C31_590.wav
+C31_594 data_thchs30/dev/C31_594.wav
+C31_596 data_thchs30/dev/C31_596.wav
+C31_605 data_thchs30/dev/C31_605.wav
+C31_616 data_thchs30/dev/C31_616.wav
+C31_640 data_thchs30/dev/C31_640.wav
+C31_661 data_thchs30/dev/C31_661.wav
+C31_663 data_thchs30/dev/C31_663.wav
+C31_668 data_thchs30/dev/C31_668.wav
+C31_672 data_thchs30/dev/C31_672.wav
+C31_675 data_thchs30/dev/C31_675.wav
+C31_678 data_thchs30/dev/C31_678.wav
+C31_683 data_thchs30/dev/C31_683.wav
+C31_693 data_thchs30/dev/C31_693.wav
+C31_710 data_thchs30/dev/C31_710.wav
+C31_735 data_thchs30/dev/C31_735.wav
+C31_747 data_thchs30/dev/C31_747.wav
+C31_749 data_thchs30/dev/C31_749.wav
+C32_516 data_thchs30/dev/C32_516.wav
+C32_524 data_thchs30/dev/C32_524.wav
+C32_533 data_thchs30/dev/C32_533.wav
+C32_543 data_thchs30/dev/C32_543.wav
+C32_558 data_thchs30/dev/C32_558.wav
+C32_570 data_thchs30/dev/C32_570.wav
+C32_577 data_thchs30/dev/C32_577.wav
+C32_583 data_thchs30/dev/C32_583.wav
+C32_590 data_thchs30/dev/C32_590.wav
+C32_598 data_thchs30/dev/C32_598.wav
+C32_604 data_thchs30/dev/C32_604.wav
+C32_632 data_thchs30/dev/C32_632.wav
+C32_637 data_thchs30/dev/C32_637.wav
+C32_650 data_thchs30/dev/C32_650.wav
+C32_652 data_thchs30/dev/C32_652.wav
+C32_667 data_thchs30/dev/C32_667.wav
+C32_678 data_thchs30/dev/C32_678.wav
+C32_710 data_thchs30/dev/C32_710.wav
+C32_713 data_thchs30/dev/C32_713.wav
+C32_742 data_thchs30/dev/C32_742.wav
+C32_746 data_thchs30/dev/C32_746.wav
+C4_515 data_thchs30/dev/C4_515.wav
+C4_520 data_thchs30/dev/C4_520.wav
+C4_542 data_thchs30/dev/C4_542.wav
+C4_543 data_thchs30/dev/C4_543.wav
+C4_544 data_thchs30/dev/C4_544.wav
+C4_574 data_thchs30/dev/C4_574.wav
+C4_575 data_thchs30/dev/C4_575.wav
+C4_584 data_thchs30/dev/C4_584.wav
+C4_599 data_thchs30/dev/C4_599.wav
+C4_622 data_thchs30/dev/C4_622.wav
+C4_623 data_thchs30/dev/C4_623.wav
+C4_652 data_thchs30/dev/C4_652.wav
+C4_661 data_thchs30/dev/C4_661.wav
+C4_666 data_thchs30/dev/C4_666.wav
+C4_683 data_thchs30/dev/C4_683.wav
+C4_713 data_thchs30/dev/C4_713.wav
+C4_715 data_thchs30/dev/C4_715.wav
+C4_733 data_thchs30/dev/C4_733.wav
+C6_525 data_thchs30/dev/C6_525.wav
+C6_531 data_thchs30/dev/C6_531.wav
+C6_533 data_thchs30/dev/C6_533.wav
+C6_545 data_thchs30/dev/C6_545.wav
+C6_560 data_thchs30/dev/C6_560.wav
+C6_564 data_thchs30/dev/C6_564.wav
+C6_582 data_thchs30/dev/C6_582.wav
+C6_596 data_thchs30/dev/C6_596.wav
+C6_598 data_thchs30/dev/C6_598.wav
+C6_601 data_thchs30/dev/C6_601.wav
+C6_618 data_thchs30/dev/C6_618.wav
+C6_644 data_thchs30/dev/C6_644.wav
+C6_653 data_thchs30/dev/C6_653.wav
+C6_679 data_thchs30/dev/C6_679.wav
+C6_681 data_thchs30/dev/C6_681.wav
+C6_682 data_thchs30/dev/C6_682.wav
+C6_707 data_thchs30/dev/C6_707.wav
+C6_714 data_thchs30/dev/C6_714.wav
+C6_715 data_thchs30/dev/C6_715.wav
+C6_735 data_thchs30/dev/C6_735.wav
+C7_512 data_thchs30/dev/C7_512.wav
+C7_520 data_thchs30/dev/C7_520.wav
+C7_541 data_thchs30/dev/C7_541.wav
+C7_557 data_thchs30/dev/C7_557.wav
+C7_596 data_thchs30/dev/C7_596.wav
+C7_613 data_thchs30/dev/C7_613.wav
+C7_617 data_thchs30/dev/C7_617.wav
+C7_631 data_thchs30/dev/C7_631.wav
+C7_638 data_thchs30/dev/C7_638.wav
+C7_648 data_thchs30/dev/C7_648.wav
+C7_662 data_thchs30/dev/C7_662.wav
+C7_679 data_thchs30/dev/C7_679.wav
+C7_693 data_thchs30/dev/C7_693.wav
+C7_700 data_thchs30/dev/C7_700.wav
+C7_701 data_thchs30/dev/C7_701.wav
+C7_702 data_thchs30/dev/C7_702.wav
+C7_704 data_thchs30/dev/C7_704.wav
+C7_711 data_thchs30/dev/C7_711.wav
+C7_747 data_thchs30/dev/C7_747.wav
+C8_513 data_thchs30/dev/C8_513.wav
+C8_517 data_thchs30/dev/C8_517.wav
+C8_527 data_thchs30/dev/C8_527.wav
+C8_537 data_thchs30/dev/C8_537.wav
+C8_548 data_thchs30/dev/C8_548.wav
+C8_559 data_thchs30/dev/C8_559.wav
+C8_569 data_thchs30/dev/C8_569.wav
+C8_594 data_thchs30/dev/C8_594.wav
+C8_624 data_thchs30/dev/C8_624.wav
+C8_628 data_thchs30/dev/C8_628.wav
+C8_644 data_thchs30/dev/C8_644.wav
+C8_655 data_thchs30/dev/C8_655.wav
+C8_657 data_thchs30/dev/C8_657.wav
+C8_672 data_thchs30/dev/C8_672.wav
+C8_684 data_thchs30/dev/C8_684.wav
+C8_697 data_thchs30/dev/C8_697.wav
+C8_747 data_thchs30/dev/C8_747.wav

datalist/thchs30/test.wav.lst

@@ -0,0 +1,2495 @@
+D11_750 data_thchs30/test/D11_750.wav
+D11_751 data_thchs30/test/D11_751.wav
+D11_752 data_thchs30/test/D11_752.wav
+D11_753 data_thchs30/test/D11_753.wav
+D11_754 data_thchs30/test/D11_754.wav
+D11_755 data_thchs30/test/D11_755.wav
+D11_756 data_thchs30/test/D11_756.wav
+D11_757 data_thchs30/test/D11_757.wav
+D11_758 data_thchs30/test/D11_758.wav
+D11_759 data_thchs30/test/D11_759.wav
+D11_760 data_thchs30/test/D11_760.wav
+D11_761 data_thchs30/test/D11_761.wav
+D11_762 data_thchs30/test/D11_762.wav
+D11_763 data_thchs30/test/D11_763.wav
+D11_764 data_thchs30/test/D11_764.wav
+D11_765 data_thchs30/test/D11_765.wav
+D11_766 data_thchs30/test/D11_766.wav
+D11_767 data_thchs30/test/D11_767.wav
+D11_768 data_thchs30/test/D11_768.wav
+D11_769 data_thchs30/test/D11_769.wav
+D11_770 data_thchs30/test/D11_770.wav
+D11_771 data_thchs30/test/D11_771.wav
+D11_772 data_thchs30/test/D11_772.wav
+D11_773 data_thchs30/test/D11_773.wav
+D11_774 data_thchs30/test/D11_774.wav
+D11_775 data_thchs30/test/D11_775.wav
+D11_776 data_thchs30/test/D11_776.wav
+D11_777 data_thchs30/test/D11_777.wav
+D11_778 data_thchs30/test/D11_778.wav
+D11_779 data_thchs30/test/D11_779.wav
+D11_780 data_thchs30/test/D11_780.wav
+D11_781 data_thchs30/test/D11_781.wav
+D11_782 data_thchs30/test/D11_782.wav
+D11_783 data_thchs30/test/D11_783.wav
+D11_785 data_thchs30/test/D11_785.wav
+D11_786 data_thchs30/test/D11_786.wav
+D11_787 data_thchs30/test/D11_787.wav
+D11_788 data_thchs30/test/D11_788.wav
+D11_789 data_thchs30/test/D11_789.wav
+D11_790 data_thchs30/test/D11_790.wav
+D11_791 data_thchs30/test/D11_791.wav
+D11_792 data_thchs30/test/D11_792.wav
+D11_793 data_thchs30/test/D11_793.wav
+D11_794 data_thchs30/test/D11_794.wav
+D11_795 data_thchs30/test/D11_795.wav
+D11_796 data_thchs30/test/D11_796.wav
+D11_797 data_thchs30/test/D11_797.wav
+D11_798 data_thchs30/test/D11_798.wav
+D11_799 data_thchs30/test/D11_799.wav
+D11_800 data_thchs30/test/D11_800.wav
+D11_801 data_thchs30/test/D11_801.wav
+D11_802 data_thchs30/test/D11_802.wav
+D11_803 data_thchs30/test/D11_803.wav
+D11_804 data_thchs30/test/D11_804.wav
+D11_805 data_thchs30/test/D11_805.wav
+D11_806 data_thchs30/test/D11_806.wav
+D11_807 data_thchs30/test/D11_807.wav
+D11_808 data_thchs30/test/D11_808.wav
+D11_809 data_thchs30/test/D11_809.wav
+D11_810 data_thchs30/test/D11_810.wav
+D11_811 data_thchs30/test/D11_811.wav
+D11_812 data_thchs30/test/D11_812.wav
+D11_813 data_thchs30/test/D11_813.wav
+D11_814 data_thchs30/test/D11_814.wav
+D11_815 data_thchs30/test/D11_815.wav
+D11_816 data_thchs30/test/D11_816.wav
+D11_817 data_thchs30/test/D11_817.wav
+D11_818 data_thchs30/test/D11_818.wav
+D11_819 data_thchs30/test/D11_819.wav
+D11_820 data_thchs30/test/D11_820.wav
+D11_821 data_thchs30/test/D11_821.wav
+D11_822 data_thchs30/test/D11_822.wav
+D11_823 data_thchs30/test/D11_823.wav
+D11_824 data_thchs30/test/D11_824.wav
+D11_825 data_thchs30/test/D11_825.wav
+D11_826 data_thchs30/test/D11_826.wav
+D11_827 data_thchs30/test/D11_827.wav
+D11_828 data_thchs30/test/D11_828.wav
+D11_829 data_thchs30/test/D11_829.wav
+D11_830 data_thchs30/test/D11_830.wav
+D11_831 data_thchs30/test/D11_831.wav
+D11_832 data_thchs30/test/D11_832.wav
+D11_833 data_thchs30/test/D11_833.wav
+D11_834 data_thchs30/test/D11_834.wav
+D11_835 data_thchs30/test/D11_835.wav
+D11_836 data_thchs30/test/D11_836.wav
+D11_837 data_thchs30/test/D11_837.wav
+D11_838 data_thchs30/test/D11_838.wav
+D11_839 data_thchs30/test/D11_839.wav
+D11_840 data_thchs30/test/D11_840.wav
+D11_841 data_thchs30/test/D11_841.wav
+D11_842 data_thchs30/test/D11_842.wav
+D11_843 data_thchs30/test/D11_843.wav
+D11_844 data_thchs30/test/D11_844.wav
+D11_845 data_thchs30/test/D11_845.wav
+D11_846 data_thchs30/test/D11_846.wav
+D11_847 data_thchs30/test/D11_847.wav
+D11_848 data_thchs30/test/D11_848.wav
+D11_849 data_thchs30/test/D11_849.wav
+D11_850 data_thchs30/test/D11_850.wav
+D11_851 data_thchs30/test/D11_851.wav
+D11_852 data_thchs30/test/D11_852.wav
+D11_853 data_thchs30/test/D11_853.wav
+D11_854 data_thchs30/test/D11_854.wav
+D11_855 data_thchs30/test/D11_855.wav
+D11_856 data_thchs30/test/D11_856.wav
+D11_857 data_thchs30/test/D11_857.wav
+D11_858 data_thchs30/test/D11_858.wav
+D11_859 data_thchs30/test/D11_859.wav
+D11_860 data_thchs30/test/D11_860.wav
+D11_861 data_thchs30/test/D11_861.wav
+D11_862 data_thchs30/test/D11_862.wav
+D11_863 data_thchs30/test/D11_863.wav
+D11_864 data_thchs30/test/D11_864.wav
+D11_865 data_thchs30/test/D11_865.wav
+D11_866 data_thchs30/test/D11_866.wav
+D11_867 data_thchs30/test/D11_867.wav
+D11_868 data_thchs30/test/D11_868.wav
+D11_869 data_thchs30/test/D11_869.wav
+D11_870 data_thchs30/test/D11_870.wav
+D11_871 data_thchs30/test/D11_871.wav
+D11_872 data_thchs30/test/D11_872.wav
+D11_873 data_thchs30/test/D11_873.wav
+D11_874 data_thchs30/test/D11_874.wav
+D11_875 data_thchs30/test/D11_875.wav
+D11_876 data_thchs30/test/D11_876.wav
+D11_877 data_thchs30/test/D11_877.wav
+D11_878 data_thchs30/test/D11_878.wav
+D11_879 data_thchs30/test/D11_879.wav
+D11_880 data_thchs30/test/D11_880.wav
+D11_881 data_thchs30/test/D11_881.wav
+D11_882 data_thchs30/test/D11_882.wav
+D11_883 data_thchs30/test/D11_883.wav
+D11_884 data_thchs30/test/D11_884.wav
+D11_885 data_thchs30/test/D11_885.wav
+D11_886 data_thchs30/test/D11_886.wav
+D11_887 data_thchs30/test/D11_887.wav
+D11_888 data_thchs30/test/D11_888.wav
+D11_889 data_thchs30/test/D11_889.wav
+D11_890 data_thchs30/test/D11_890.wav
+D11_891 data_thchs30/test/D11_891.wav
+D11_892 data_thchs30/test/D11_892.wav
+D11_893 data_thchs30/test/D11_893.wav
+D11_894 data_thchs30/test/D11_894.wav
+D11_895 data_thchs30/test/D11_895.wav
+D11_896 data_thchs30/test/D11_896.wav
+D11_897 data_thchs30/test/D11_897.wav
+D11_898 data_thchs30/test/D11_898.wav
+D11_899 data_thchs30/test/D11_899.wav
+D11_900 data_thchs30/test/D11_900.wav
+D11_901 data_thchs30/test/D11_901.wav
+D11_902 data_thchs30/test/D11_902.wav
+D11_903 data_thchs30/test/D11_903.wav
+D11_904 data_thchs30/test/D11_904.wav
+D11_905 data_thchs30/test/D11_905.wav
+D11_906 data_thchs30/test/D11_906.wav
+D11_907 data_thchs30/test/D11_907.wav
+D11_908 data_thchs30/test/D11_908.wav
+D11_909 data_thchs30/test/D11_909.wav
+D11_910 data_thchs30/test/D11_910.wav
+D11_911 data_thchs30/test/D11_911.wav
+D11_912 data_thchs30/test/D11_912.wav
+D11_913 data_thchs30/test/D11_913.wav
+D11_914 data_thchs30/test/D11_914.wav
+D11_915 data_thchs30/test/D11_915.wav
+D11_916 data_thchs30/test/D11_916.wav
+D11_917 data_thchs30/test/D11_917.wav
+D11_918 data_thchs30/test/D11_918.wav
+D11_919 data_thchs30/test/D11_919.wav
+D11_920 data_thchs30/test/D11_920.wav
+D11_921 data_thchs30/test/D11_921.wav
+D11_922 data_thchs30/test/D11_922.wav
+D11_923 data_thchs30/test/D11_923.wav
+D11_924 data_thchs30/test/D11_924.wav
+D11_925 data_thchs30/test/D11_925.wav
+D11_926 data_thchs30/test/D11_926.wav
+D11_927 data_thchs30/test/D11_927.wav
+D11_928 data_thchs30/test/D11_928.wav
+D11_929 data_thchs30/test/D11_929.wav
+D11_930 data_thchs30/test/D11_930.wav
+D11_931 data_thchs30/test/D11_931.wav
+D11_932 data_thchs30/test/D11_932.wav
+D11_933 data_thchs30/test/D11_933.wav
+D11_934 data_thchs30/test/D11_934.wav
+D11_935 data_thchs30/test/D11_935.wav
+D11_936 data_thchs30/test/D11_936.wav
+D11_937 data_thchs30/test/D11_937.wav
+D11_938 data_thchs30/test/D11_938.wav
+D11_939 data_thchs30/test/D11_939.wav
+D11_940 data_thchs30/test/D11_940.wav
+D11_941 data_thchs30/test/D11_941.wav
+D11_942 data_thchs30/test/D11_942.wav
+D11_943 data_thchs30/test/D11_943.wav
+D11_944 data_thchs30/test/D11_944.wav
+D11_945 data_thchs30/test/D11_945.wav
+D11_946 data_thchs30/test/D11_946.wav
+D11_947 data_thchs30/test/D11_947.wav
+D11_948 data_thchs30/test/D11_948.wav
+D11_949 data_thchs30/test/D11_949.wav
+D11_950 data_thchs30/test/D11_950.wav
+D11_951 data_thchs30/test/D11_951.wav
+D11_952 data_thchs30/test/D11_952.wav
+D11_953 data_thchs30/test/D11_953.wav
+D11_954 data_thchs30/test/D11_954.wav
+D11_955 data_thchs30/test/D11_955.wav
+D11_956 data_thchs30/test/D11_956.wav
+D11_957 data_thchs30/test/D11_957.wav
+D11_958 data_thchs30/test/D11_958.wav
+D11_959 data_thchs30/test/D11_959.wav
+D11_960 data_thchs30/test/D11_960.wav
+D11_961 data_thchs30/test/D11_961.wav
+D11_962 data_thchs30/test/D11_962.wav
+D11_963 data_thchs30/test/D11_963.wav
+D11_964 data_thchs30/test/D11_964.wav
+D11_965 data_thchs30/test/D11_965.wav
+D11_966 data_thchs30/test/D11_966.wav
+D11_967 data_thchs30/test/D11_967.wav
+D11_968 data_thchs30/test/D11_968.wav
+D11_969 data_thchs30/test/D11_969.wav
+D11_970 data_thchs30/test/D11_970.wav
+D11_971 data_thchs30/test/D11_971.wav
+D11_972 data_thchs30/test/D11_972.wav
+D11_973 data_thchs30/test/D11_973.wav
+D11_974 data_thchs30/test/D11_974.wav
+D11_975 data_thchs30/test/D11_975.wav
+D11_976 data_thchs30/test/D11_976.wav
+D11_977 data_thchs30/test/D11_977.wav
+D11_978 data_thchs30/test/D11_978.wav
+D11_979 data_thchs30/test/D11_979.wav
+D11_980 data_thchs30/test/D11_980.wav
+D11_981 data_thchs30/test/D11_981.wav
+D11_982 data_thchs30/test/D11_982.wav
+D11_983 data_thchs30/test/D11_983.wav
+D11_984 data_thchs30/test/D11_984.wav
+D11_985 data_thchs30/test/D11_985.wav
+D11_986 data_thchs30/test/D11_986.wav
+D11_987 data_thchs30/test/D11_987.wav
+D11_988 data_thchs30/test/D11_988.wav
+D11_989 data_thchs30/test/D11_989.wav
+D11_990 data_thchs30/test/D11_990.wav
+D11_991 data_thchs30/test/D11_991.wav
+D11_992 data_thchs30/test/D11_992.wav
+D11_993 data_thchs30/test/D11_993.wav
+D11_994 data_thchs30/test/D11_994.wav
+D11_995 data_thchs30/test/D11_995.wav
+D11_996 data_thchs30/test/D11_996.wav
+D11_997 data_thchs30/test/D11_997.wav
+D11_998 data_thchs30/test/D11_998.wav
+D11_999 data_thchs30/test/D11_999.wav
+D12_750 data_thchs30/test/D12_750.wav
+D12_751 data_thchs30/test/D12_751.wav
+D12_752 data_thchs30/test/D12_752.wav
+D12_753 data_thchs30/test/D12_753.wav
+D12_754 data_thchs30/test/D12_754.wav
+D12_755 data_thchs30/test/D12_755.wav
+D12_756 data_thchs30/test/D12_756.wav
+D12_757 data_thchs30/test/D12_757.wav
+D12_758 data_thchs30/test/D12_758.wav
+D12_759 data_thchs30/test/D12_759.wav
+D12_760 data_thchs30/test/D12_760.wav
+D12_761 data_thchs30/test/D12_761.wav
+D12_762 data_thchs30/test/D12_762.wav
+D12_763 data_thchs30/test/D12_763.wav
+D12_764 data_thchs30/test/D12_764.wav
+D12_765 data_thchs30/test/D12_765.wav
+D12_766 data_thchs30/test/D12_766.wav
+D12_767 data_thchs30/test/D12_767.wav
+D12_768 data_thchs30/test/D12_768.wav
+D12_769 data_thchs30/test/D12_769.wav
+D12_770 data_thchs30/test/D12_770.wav
+D12_771 data_thchs30/test/D12_771.wav
+D12_772 data_thchs30/test/D12_772.wav
+D12_773 data_thchs30/test/D12_773.wav
+D12_774 data_thchs30/test/D12_774.wav
+D12_775 data_thchs30/test/D12_775.wav
+D12_776 data_thchs30/test/D12_776.wav
+D12_777 data_thchs30/test/D12_777.wav
+D12_778 data_thchs30/test/D12_778.wav
+D12_779 data_thchs30/test/D12_779.wav
+D12_780 data_thchs30/test/D12_780.wav
+D12_781 data_thchs30/test/D12_781.wav
+D12_782 data_thchs30/test/D12_782.wav
+D12_783 data_thchs30/test/D12_783.wav
+D12_784 data_thchs30/test/D12_784.wav
+D12_785 data_thchs30/test/D12_785.wav
+D12_786 data_thchs30/test/D12_786.wav
+D12_787 data_thchs30/test/D12_787.wav
+D12_788 data_thchs30/test/D12_788.wav
+D12_789 data_thchs30/test/D12_789.wav
+D12_790 data_thchs30/test/D12_790.wav
+D12_791 data_thchs30/test/D12_791.wav
+D12_792 data_thchs30/test/D12_792.wav
+D12_793 data_thchs30/test/D12_793.wav
+D12_794 data_thchs30/test/D12_794.wav
+D12_795 data_thchs30/test/D12_795.wav
+D12_796 data_thchs30/test/D12_796.wav
+D12_797 data_thchs30/test/D12_797.wav
+D12_798 data_thchs30/test/D12_798.wav
+D12_799 data_thchs30/test/D12_799.wav
+D12_800 data_thchs30/test/D12_800.wav
+D12_801 data_thchs30/test/D12_801.wav
+D12_802 data_thchs30/test/D12_802.wav
+D12_803 data_thchs30/test/D12_803.wav
+D12_804 data_thchs30/test/D12_804.wav
+D12_805 data_thchs30/test/D12_805.wav
+D12_806 data_thchs30/test/D12_806.wav
+D12_809 data_thchs30/test/D12_809.wav
+D12_810 data_thchs30/test/D12_810.wav
+D12_811 data_thchs30/test/D12_811.wav
+D12_812 data_thchs30/test/D12_812.wav
+D12_813 data_thchs30/test/D12_813.wav
+D12_814 data_thchs30/test/D12_814.wav
+D12_815 data_thchs30/test/D12_815.wav
+D12_816 data_thchs30/test/D12_816.wav
+D12_817 data_thchs30/test/D12_817.wav
+D12_818 data_thchs30/test/D12_818.wav
+D12_819 data_thchs30/test/D12_819.wav
+D12_820 data_thchs30/test/D12_820.wav
+D12_821 data_thchs30/test/D12_821.wav
+D12_822 data_thchs30/test/D12_822.wav
+D12_823 data_thchs30/test/D12_823.wav
+D12_824 data_thchs30/test/D12_824.wav
+D12_825 data_thchs30/test/D12_825.wav
+D12_826 data_thchs30/test/D12_826.wav
+D12_827 data_thchs30/test/D12_827.wav
+D12_828 data_thchs30/test/D12_828.wav
+D12_829 data_thchs30/test/D12_829.wav
+D12_830 data_thchs30/test/D12_830.wav
+D12_831 data_thchs30/test/D12_831.wav
+D12_832 data_thchs30/test/D12_832.wav
+D12_833 data_thchs30/test/D12_833.wav
+D12_834 data_thchs30/test/D12_834.wav
+D12_835 data_thchs30/test/D12_835.wav
+D12_836 data_thchs30/test/D12_836.wav
+D12_837 data_thchs30/test/D12_837.wav
+D12_838 data_thchs30/test/D12_838.wav
+D12_839 data_thchs30/test/D12_839.wav
+D12_840 data_thchs30/test/D12_840.wav
+D12_841 data_thchs30/test/D12_841.wav
+D12_842 data_thchs30/test/D12_842.wav
+D12_843 data_thchs30/test/D12_843.wav
+D12_844 data_thchs30/test/D12_844.wav
+D12_845 data_thchs30/test/D12_845.wav
+D12_846 data_thchs30/test/D12_846.wav
+D12_847 data_thchs30/test/D12_847.wav
+D12_848 data_thchs30/test/D12_848.wav
+D12_849 data_thchs30/test/D12_849.wav
+D12_850 data_thchs30/test/D12_850.wav
+D12_851 data_thchs30/test/D12_851.wav
+D12_852 data_thchs30/test/D12_852.wav
+D12_853 data_thchs30/test/D12_853.wav
+D12_854 data_thchs30/test/D12_854.wav
+D12_855 data_thchs30/test/D12_855.wav
+D12_856 data_thchs30/test/D12_856.wav
+D12_857 data_thchs30/test/D12_857.wav
+D12_858 data_thchs30/test/D12_858.wav
+D12_859 data_thchs30/test/D12_859.wav
+D12_860 data_thchs30/test/D12_860.wav
+D12_861 data_thchs30/test/D12_861.wav
+D12_862 data_thchs30/test/D12_862.wav
+D12_863 data_thchs30/test/D12_863.wav
+D12_864 data_thchs30/test/D12_864.wav
+D12_865 data_thchs30/test/D12_865.wav
+D12_866 data_thchs30/test/D12_866.wav
+D12_867 data_thchs30/test/D12_867.wav
+D12_868 data_thchs30/test/D12_868.wav
+D12_869 data_thchs30/test/D12_869.wav
+D12_870 data_thchs30/test/D12_870.wav
+D12_871 data_thchs30/test/D12_871.wav
+D12_872 data_thchs30/test/D12_872.wav
+D12_873 data_thchs30/test/D12_873.wav
+D12_874 data_thchs30/test/D12_874.wav
+D12_875 data_thchs30/test/D12_875.wav
+D12_876 data_thchs30/test/D12_876.wav
+D12_877 data_thchs30/test/D12_877.wav
+D12_878 data_thchs30/test/D12_878.wav
+D12_879 data_thchs30/test/D12_879.wav
+D12_880 data_thchs30/test/D12_880.wav
+D12_881 data_thchs30/test/D12_881.wav
+D12_882 data_thchs30/test/D12_882.wav
+D12_883 data_thchs30/test/D12_883.wav
+D12_884 data_thchs30/test/D12_884.wav
+D12_885 data_thchs30/test/D12_885.wav
+D12_886 data_thchs30/test/D12_886.wav
+D12_887 data_thchs30/test/D12_887.wav
+D12_888 data_thchs30/test/D12_888.wav
+D12_889 data_thchs30/test/D12_889.wav
+D12_890 data_thchs30/test/D12_890.wav
+D12_891 data_thchs30/test/D12_891.wav
+D12_892 data_thchs30/test/D12_892.wav
+D12_893 data_thchs30/test/D12_893.wav
+D12_894 data_thchs30/test/D12_894.wav
+D12_895 data_thchs30/test/D12_895.wav
+D12_896 data_thchs30/test/D12_896.wav
+D12_897 data_thchs30/test/D12_897.wav
+D12_898 data_thchs30/test/D12_898.wav
+D12_899 data_thchs30/test/D12_899.wav
+D12_900 data_thchs30/test/D12_900.wav
+D12_901 data_thchs30/test/D12_901.wav
+D12_902 data_thchs30/test/D12_902.wav
+D12_903 data_thchs30/test/D12_903.wav
+D12_904 data_thchs30/test/D12_904.wav
+D12_905 data_thchs30/test/D12_905.wav
+D12_906 data_thchs30/test/D12_906.wav
+D12_907 data_thchs30/test/D12_907.wav
+D12_908 data_thchs30/test/D12_908.wav
+D12_909 data_thchs30/test/D12_909.wav
+D12_910 data_thchs30/test/D12_910.wav
+D12_911 data_thchs30/test/D12_911.wav
+D12_912 data_thchs30/test/D12_912.wav
+D12_913 data_thchs30/test/D12_913.wav
+D12_914 data_thchs30/test/D12_914.wav
+D12_915 data_thchs30/test/D12_915.wav
+D12_916 data_thchs30/test/D12_916.wav
+D12_917 data_thchs30/test/D12_917.wav
+D12_918 data_thchs30/test/D12_918.wav
+D12_919 data_thchs30/test/D12_919.wav
+D12_920 data_thchs30/test/D12_920.wav
+D12_921 data_thchs30/test/D12_921.wav
+D12_922 data_thchs30/test/D12_922.wav
+D12_923 data_thchs30/test/D12_923.wav
+D12_924 data_thchs30/test/D12_924.wav
+D12_925 data_thchs30/test/D12_925.wav
+D12_926 data_thchs30/test/D12_926.wav
+D12_927 data_thchs30/test/D12_927.wav
+D12_928 data_thchs30/test/D12_928.wav
+D12_929 data_thchs30/test/D12_929.wav
+D12_930 data_thchs30/test/D12_930.wav
+D12_931 data_thchs30/test/D12_931.wav
+D12_932 data_thchs30/test/D12_932.wav
+D12_933 data_thchs30/test/D12_933.wav
+D12_934 data_thchs30/test/D12_934.wav
+D12_935 data_thchs30/test/D12_935.wav
+D12_936 data_thchs30/test/D12_936.wav
+D12_937 data_thchs30/test/D12_937.wav
+D12_938 data_thchs30/test/D12_938.wav
+D12_939 data_thchs30/test/D12_939.wav
+D12_940 data_thchs30/test/D12_940.wav
+D12_941 data_thchs30/test/D12_941.wav
+D12_942 data_thchs30/test/D12_942.wav
+D12_943 data_thchs30/test/D12_943.wav
+D12_944 data_thchs30/test/D12_944.wav
+D12_945 data_thchs30/test/D12_945.wav
+D12_946 data_thchs30/test/D12_946.wav
+D12_947 data_thchs30/test/D12_947.wav
+D12_948 data_thchs30/test/D12_948.wav
+D12_949 data_thchs30/test/D12_949.wav
+D12_950 data_thchs30/test/D12_950.wav
+D12_951 data_thchs30/test/D12_951.wav
+D12_952 data_thchs30/test/D12_952.wav
+D12_953 data_thchs30/test/D12_953.wav
+D12_954 data_thchs30/test/D12_954.wav
+D12_955 data_thchs30/test/D12_955.wav
+D12_956 data_thchs30/test/D12_956.wav
+D12_957 data_thchs30/test/D12_957.wav
+D12_958 data_thchs30/test/D12_958.wav
+D12_959 data_thchs30/test/D12_959.wav
+D12_960 data_thchs30/test/D12_960.wav
+D12_961 data_thchs30/test/D12_961.wav
+D12_962 data_thchs30/test/D12_962.wav
+D12_963 data_thchs30/test/D12_963.wav
+D12_964 data_thchs30/test/D12_964.wav
+D12_965 data_thchs30/test/D12_965.wav
+D12_966 data_thchs30/test/D12_966.wav
+D12_967 data_thchs30/test/D12_967.wav
+D12_968 data_thchs30/test/D12_968.wav
+D12_969 data_thchs30/test/D12_969.wav
+D12_970 data_thchs30/test/D12_970.wav
+D12_971 data_thchs30/test/D12_971.wav
+D12_972 data_thchs30/test/D12_972.wav
+D12_973 data_thchs30/test/D12_973.wav
+D12_974 data_thchs30/test/D12_974.wav
+D12_975 data_thchs30/test/D12_975.wav
+D12_976 data_thchs30/test/D12_976.wav
+D12_977 data_thchs30/test/D12_977.wav
+D12_978 data_thchs30/test/D12_978.wav
+D12_979 data_thchs30/test/D12_979.wav
+D12_980 data_thchs30/test/D12_980.wav
+D12_981 data_thchs30/test/D12_981.wav
+D12_982 data_thchs30/test/D12_982.wav
+D12_983 data_thchs30/test/D12_983.wav
+D12_984 data_thchs30/test/D12_984.wav
+D12_985 data_thchs30/test/D12_985.wav
+D12_986 data_thchs30/test/D12_986.wav
+D12_987 data_thchs30/test/D12_987.wav
+D12_988 data_thchs30/test/D12_988.wav
+D12_989 data_thchs30/test/D12_989.wav
+D12_990 data_thchs30/test/D12_990.wav
+D12_991 data_thchs30/test/D12_991.wav
+D12_992 data_thchs30/test/D12_992.wav
+D12_993 data_thchs30/test/D12_993.wav
+D12_994 data_thchs30/test/D12_994.wav
+D12_995 data_thchs30/test/D12_995.wav
+D12_996 data_thchs30/test/D12_996.wav
+D12_997 data_thchs30/test/D12_997.wav
+D12_998 data_thchs30/test/D12_998.wav
+D12_999 data_thchs30/test/D12_999.wav
+D13_750 data_thchs30/test/D13_750.wav
+D13_751 data_thchs30/test/D13_751.wav
+D13_752 data_thchs30/test/D13_752.wav
+D13_753 data_thchs30/test/D13_753.wav
+D13_754 data_thchs30/test/D13_754.wav
+D13_755 data_thchs30/test/D13_755.wav
+D13_756 data_thchs30/test/D13_756.wav
+D13_757 data_thchs30/test/D13_757.wav
+D13_758 data_thchs30/test/D13_758.wav
+D13_759 data_thchs30/test/D13_759.wav
+D13_760 data_thchs30/test/D13_760.wav
+D13_761 data_thchs30/test/D13_761.wav
+D13_762 data_thchs30/test/D13_762.wav
+D13_763 data_thchs30/test/D13_763.wav
+D13_764 data_thchs30/test/D13_764.wav
+D13_765 data_thchs30/test/D13_765.wav
+D13_766 data_thchs30/test/D13_766.wav
+D13_767 data_thchs30/test/D13_767.wav
+D13_768 data_thchs30/test/D13_768.wav
+D13_769 data_thchs30/test/D13_769.wav
+D13_770 data_thchs30/test/D13_770.wav
+D13_771 data_thchs30/test/D13_771.wav
+D13_772 data_thchs30/test/D13_772.wav
+D13_773 data_thchs30/test/D13_773.wav
+D13_774 data_thchs30/test/D13_774.wav
+D13_775 data_thchs30/test/D13_775.wav
+D13_776 data_thchs30/test/D13_776.wav
+D13_777 data_thchs30/test/D13_777.wav
+D13_778 data_thchs30/test/D13_778.wav
+D13_779 data_thchs30/test/D13_779.wav
+D13_780 data_thchs30/test/D13_780.wav
+D13_781 data_thchs30/test/D13_781.wav
+D13_782 data_thchs30/test/D13_782.wav
+D13_783 data_thchs30/test/D13_783.wav
+D13_784 data_thchs30/test/D13_784.wav
+D13_785 data_thchs30/test/D13_785.wav
+D13_786 data_thchs30/test/D13_786.wav
+D13_787 data_thchs30/test/D13_787.wav
+D13_788 data_thchs30/test/D13_788.wav
+D13_789 data_thchs30/test/D13_789.wav
+D13_790 data_thchs30/test/D13_790.wav
+D13_791 data_thchs30/test/D13_791.wav
+D13_792 data_thchs30/test/D13_792.wav
+D13_793 data_thchs30/test/D13_793.wav
+D13_794 data_thchs30/test/D13_794.wav
+D13_795 data_thchs30/test/D13_795.wav
+D13_796 data_thchs30/test/D13_796.wav
+D13_797 data_thchs30/test/D13_797.wav
+D13_798 data_thchs30/test/D13_798.wav
+D13_799 data_thchs30/test/D13_799.wav
+D13_800 data_thchs30/test/D13_800.wav
+D13_801 data_thchs30/test/D13_801.wav
+D13_802 data_thchs30/test/D13_802.wav
+D13_803 data_thchs30/test/D13_803.wav
+D13_804 data_thchs30/test/D13_804.wav
+D13_805 data_thchs30/test/D13_805.wav
+D13_806 data_thchs30/test/D13_806.wav
+D13_807 data_thchs30/test/D13_807.wav
+D13_808 data_thchs30/test/D13_808.wav
+D13_809 data_thchs30/test/D13_809.wav
+D13_810 data_thchs30/test/D13_810.wav
+D13_811 data_thchs30/test/D13_811.wav
+D13_812 data_thchs30/test/D13_812.wav
+D13_813 data_thchs30/test/D13_813.wav
+D13_814 data_thchs30/test/D13_814.wav
+D13_815 data_thchs30/test/D13_815.wav
+D13_816 data_thchs30/test/D13_816.wav
+D13_817 data_thchs30/test/D13_817.wav
+D13_818 data_thchs30/test/D13_818.wav
+D13_819 data_thchs30/test/D13_819.wav
+D13_820 data_thchs30/test/D13_820.wav
+D13_821 data_thchs30/test/D13_821.wav
+D13_822 data_thchs30/test/D13_822.wav
+D13_823 data_thchs30/test/D13_823.wav
+D13_824 data_thchs30/test/D13_824.wav
+D13_825 data_thchs30/test/D13_825.wav
+D13_826 data_thchs30/test/D13_826.wav
+D13_827 data_thchs30/test/D13_827.wav
+D13_828 data_thchs30/test/D13_828.wav
+D13_829 data_thchs30/test/D13_829.wav
+D13_830 data_thchs30/test/D13_830.wav
+D13_831 data_thchs30/test/D13_831.wav
+D13_832 data_thchs30/test/D13_832.wav
+D13_833 data_thchs30/test/D13_833.wav
+D13_834 data_thchs30/test/D13_834.wav
+D13_835 data_thchs30/test/D13_835.wav
+D13_836 data_thchs30/test/D13_836.wav
+D13_837 data_thchs30/test/D13_837.wav
+D13_838 data_thchs30/test/D13_838.wav
+D13_839 data_thchs30/test/D13_839.wav
+D13_840 data_thchs30/test/D13_840.wav
+D13_841 data_thchs30/test/D13_841.wav
+D13_842 data_thchs30/test/D13_842.wav
+D13_843 data_thchs30/test/D13_843.wav
+D13_844 data_thchs30/test/D13_844.wav
+D13_845 data_thchs30/test/D13_845.wav
+D13_846 data_thchs30/test/D13_846.wav
+D13_847 data_thchs30/test/D13_847.wav
+D13_848 data_thchs30/test/D13_848.wav
+D13_849 data_thchs30/test/D13_849.wav
+D13_850 data_thchs30/test/D13_850.wav
+D13_851 data_thchs30/test/D13_851.wav
+D13_852 data_thchs30/test/D13_852.wav
+D13_853 data_thchs30/test/D13_853.wav
+D13_854 data_thchs30/test/D13_854.wav
+D13_855 data_thchs30/test/D13_855.wav
+D13_856 data_thchs30/test/D13_856.wav
+D13_857 data_thchs30/test/D13_857.wav
+D13_858 data_thchs30/test/D13_858.wav
+D13_859 data_thchs30/test/D13_859.wav
+D13_860 data_thchs30/test/D13_860.wav
+D13_861 data_thchs30/test/D13_861.wav
+D13_862 data_thchs30/test/D13_862.wav
+D13_863 data_thchs30/test/D13_863.wav
+D13_864 data_thchs30/test/D13_864.wav
+D13_865 data_thchs30/test/D13_865.wav
+D13_866 data_thchs30/test/D13_866.wav
+D13_867 data_thchs30/test/D13_867.wav
+D13_868 data_thchs30/test/D13_868.wav
+D13_869 data_thchs30/test/D13_869.wav
+D13_870 data_thchs30/test/D13_870.wav
+D13_871 data_thchs30/test/D13_871.wav
+D13_872 data_thchs30/test/D13_872.wav
+D13_873 data_thchs30/test/D13_873.wav
+D13_874 data_thchs30/test/D13_874.wav
+D13_875 data_thchs30/test/D13_875.wav
+D13_876 data_thchs30/test/D13_876.wav
+D13_877 data_thchs30/test/D13_877.wav
+D13_878 data_thchs30/test/D13_878.wav
+D13_879 data_thchs30/test/D13_879.wav
+D13_880 data_thchs30/test/D13_880.wav
+D13_881 data_thchs30/test/D13_881.wav
+D13_882 data_thchs30/test/D13_882.wav
+D13_883 data_thchs30/test/D13_883.wav
+D13_884 data_thchs30/test/D13_884.wav
+D13_885 data_thchs30/test/D13_885.wav
+D13_886 data_thchs30/test/D13_886.wav
+D13_887 data_thchs30/test/D13_887.wav
+D13_888 data_thchs30/test/D13_888.wav
+D13_889 data_thchs30/test/D13_889.wav
+D13_890 data_thchs30/test/D13_890.wav
+D13_891 data_thchs30/test/D13_891.wav
+D13_892 data_thchs30/test/D13_892.wav
+D13_893 data_thchs30/test/D13_893.wav
+D13_894 data_thchs30/test/D13_894.wav
+D13_895 data_thchs30/test/D13_895.wav
+D13_896 data_thchs30/test/D13_896.wav
+D13_897 data_thchs30/test/D13_897.wav
+D13_898 data_thchs30/test/D13_898.wav
+D13_899 data_thchs30/test/D13_899.wav
+D13_900 data_thchs30/test/D13_900.wav
+D13_901 data_thchs30/test/D13_901.wav
+D13_902 data_thchs30/test/D13_902.wav
+D13_903 data_thchs30/test/D13_903.wav
+D13_904 data_thchs30/test/D13_904.wav
+D13_905 data_thchs30/test/D13_905.wav
+D13_906 data_thchs30/test/D13_906.wav
+D13_907 data_thchs30/test/D13_907.wav
+D13_908 data_thchs30/test/D13_908.wav
+D13_909 data_thchs30/test/D13_909.wav
+D13_910 data_thchs30/test/D13_910.wav
+D13_911 data_thchs30/test/D13_911.wav
+D13_912 data_thchs30/test/D13_912.wav
+D13_913 data_thchs30/test/D13_913.wav
+D13_914 data_thchs30/test/D13_914.wav
+D13_915 data_thchs30/test/D13_915.wav
+D13_916 data_thchs30/test/D13_916.wav
+D13_917 data_thchs30/test/D13_917.wav
+D13_918 data_thchs30/test/D13_918.wav
+D13_919 data_thchs30/test/D13_919.wav
+D13_920 data_thchs30/test/D13_920.wav
+D13_921 data_thchs30/test/D13_921.wav
+D13_922 data_thchs30/test/D13_922.wav
+D13_923 data_thchs30/test/D13_923.wav
+D13_924 data_thchs30/test/D13_924.wav
+D13_925 data_thchs30/test/D13_925.wav
+D13_926 data_thchs30/test/D13_926.wav
+D13_927 data_thchs30/test/D13_927.wav
+D13_928 data_thchs30/test/D13_928.wav
+D13_929 data_thchs30/test/D13_929.wav
+D13_930 data_thchs30/test/D13_930.wav
+D13_931 data_thchs30/test/D13_931.wav
+D13_932 data_thchs30/test/D13_932.wav
+D13_933 data_thchs30/test/D13_933.wav
+D13_934 data_thchs30/test/D13_934.wav
+D13_935 data_thchs30/test/D13_935.wav
+D13_936 data_thchs30/test/D13_936.wav
+D13_937 data_thchs30/test/D13_937.wav
+D13_938 data_thchs30/test/D13_938.wav
+D13_939 data_thchs30/test/D13_939.wav
+D13_940 data_thchs30/test/D13_940.wav
+D13_941 data_thchs30/test/D13_941.wav
+D13_942 data_thchs30/test/D13_942.wav
+D13_943 data_thchs30/test/D13_943.wav
+D13_944 data_thchs30/test/D13_944.wav
+D13_945 data_thchs30/test/D13_945.wav
+D13_946 data_thchs30/test/D13_946.wav
+D13_947 data_thchs30/test/D13_947.wav
+D13_948 data_thchs30/test/D13_948.wav
+D13_949 data_thchs30/test/D13_949.wav
+D13_950 data_thchs30/test/D13_950.wav
+D13_951 data_thchs30/test/D13_951.wav
+D13_952 data_thchs30/test/D13_952.wav
+D13_953 data_thchs30/test/D13_953.wav
+D13_954 data_thchs30/test/D13_954.wav
+D13_955 data_thchs30/test/D13_955.wav
+D13_956 data_thchs30/test/D13_956.wav
+D13_957 data_thchs30/test/D13_957.wav
+D13_958 data_thchs30/test/D13_958.wav
+D13_959 data_thchs30/test/D13_959.wav
+D13_960 data_thchs30/test/D13_960.wav
+D13_961 data_thchs30/test/D13_961.wav
+D13_962 data_thchs30/test/D13_962.wav
+D13_963 data_thchs30/test/D13_963.wav
+D13_964 data_thchs30/test/D13_964.wav
+D13_965 data_thchs30/test/D13_965.wav
+D13_966 data_thchs30/test/D13_966.wav
+D13_967 data_thchs30/test/D13_967.wav
+D13_968 data_thchs30/test/D13_968.wav
+D13_969 data_thchs30/test/D13_969.wav
+D13_970 data_thchs30/test/D13_970.wav
+D13_971 data_thchs30/test/D13_971.wav
+D13_972 data_thchs30/test/D13_972.wav
+D13_973 data_thchs30/test/D13_973.wav
+D13_974 data_thchs30/test/D13_974.wav
+D13_975 data_thchs30/test/D13_975.wav
+D13_976 data_thchs30/test/D13_976.wav
+D13_977 data_thchs30/test/D13_977.wav
+D13_978 data_thchs30/test/D13_978.wav
+D13_979 data_thchs30/test/D13_979.wav
+D13_980 data_thchs30/test/D13_980.wav
+D13_981 data_thchs30/test/D13_981.wav
+D13_982 data_thchs30/test/D13_982.wav
+D13_983 data_thchs30/test/D13_983.wav
+D13_984 data_thchs30/test/D13_984.wav
+D13_985 data_thchs30/test/D13_985.wav
+D13_986 data_thchs30/test/D13_986.wav
+D13_987 data_thchs30/test/D13_987.wav
+D13_988 data_thchs30/test/D13_988.wav
+D13_989 data_thchs30/test/D13_989.wav
+D13_990 data_thchs30/test/D13_990.wav
+D13_991 data_thchs30/test/D13_991.wav
+D13_992 data_thchs30/test/D13_992.wav
+D13_993 data_thchs30/test/D13_993.wav
+D13_994 data_thchs30/test/D13_994.wav
+D13_995 data_thchs30/test/D13_995.wav
+D13_996 data_thchs30/test/D13_996.wav
+D13_997 data_thchs30/test/D13_997.wav
+D13_998 data_thchs30/test/D13_998.wav
+D13_999 data_thchs30/test/D13_999.wav
+D21_750 data_thchs30/test/D21_750.wav
+D21_751 data_thchs30/test/D21_751.wav
+D21_752 data_thchs30/test/D21_752.wav
+D21_753 data_thchs30/test/D21_753.wav
+D21_754 data_thchs30/test/D21_754.wav
+D21_755 data_thchs30/test/D21_755.wav
+D21_756 data_thchs30/test/D21_756.wav
+D21_757 data_thchs30/test/D21_757.wav
+D21_758 data_thchs30/test/D21_758.wav
+D21_759 data_thchs30/test/D21_759.wav
+D21_760 data_thchs30/test/D21_760.wav
+D21_761 data_thchs30/test/D21_761.wav
+D21_762 data_thchs30/test/D21_762.wav
+D21_763 data_thchs30/test/D21_763.wav
+D21_764 data_thchs30/test/D21_764.wav
+D21_765 data_thchs30/test/D21_765.wav
+D21_766 data_thchs30/test/D21_766.wav
+D21_767 data_thchs30/test/D21_767.wav
+D21_768 data_thchs30/test/D21_768.wav
+D21_769 data_thchs30/test/D21_769.wav
+D21_770 data_thchs30/test/D21_770.wav
+D21_771 data_thchs30/test/D21_771.wav
+D21_772 data_thchs30/test/D21_772.wav
+D21_773 data_thchs30/test/D21_773.wav
+D21_774 data_thchs30/test/D21_774.wav
+D21_775 data_thchs30/test/D21_775.wav
+D21_776 data_thchs30/test/D21_776.wav
+D21_777 data_thchs30/test/D21_777.wav
+D21_778 data_thchs30/test/D21_778.wav
+D21_779 data_thchs30/test/D21_779.wav
+D21_780 data_thchs30/test/D21_780.wav
+D21_781 data_thchs30/test/D21_781.wav
+D21_782 data_thchs30/test/D21_782.wav
+D21_783 data_thchs30/test/D21_783.wav
+D21_784 data_thchs30/test/D21_784.wav
+D21_785 data_thchs30/test/D21_785.wav
+D21_786 data_thchs30/test/D21_786.wav
+D21_787 data_thchs30/test/D21_787.wav
+D21_788 data_thchs30/test/D21_788.wav
+D21_789 data_thchs30/test/D21_789.wav
+D21_790 data_thchs30/test/D21_790.wav
+D21_791 data_thchs30/test/D21_791.wav
+D21_792 data_thchs30/test/D21_792.wav
+D21_793 data_thchs30/test/D21_793.wav
+D21_794 data_thchs30/test/D21_794.wav
+D21_795 data_thchs30/test/D21_795.wav
+D21_796 data_thchs30/test/D21_796.wav
+D21_797 data_thchs30/test/D21_797.wav
+D21_798 data_thchs30/test/D21_798.wav
+D21_799 data_thchs30/test/D21_799.wav
+D21_800 data_thchs30/test/D21_800.wav
+D21_801 data_thchs30/test/D21_801.wav
+D21_802 data_thchs30/test/D21_802.wav
+D21_803 data_thchs30/test/D21_803.wav
+D21_804 data_thchs30/test/D21_804.wav
+D21_805 data_thchs30/test/D21_805.wav
+D21_806 data_thchs30/test/D21_806.wav
+D21_807 data_thchs30/test/D21_807.wav
+D21_808 data_thchs30/test/D21_808.wav
+D21_809 data_thchs30/test/D21_809.wav
+D21_810 data_thchs30/test/D21_810.wav
+D21_811 data_thchs30/test/D21_811.wav
+D21_812 data_thchs30/test/D21_812.wav
+D21_813 data_thchs30/test/D21_813.wav
+D21_814 data_thchs30/test/D21_814.wav
+D21_815 data_thchs30/test/D21_815.wav
+D21_816 data_thchs30/test/D21_816.wav
+D21_817 data_thchs30/test/D21_817.wav
+D21_818 data_thchs30/test/D21_818.wav
+D21_819 data_thchs30/test/D21_819.wav
+D21_820 data_thchs30/test/D21_820.wav
+D21_821 data_thchs30/test/D21_821.wav
+D21_822 data_thchs30/test/D21_822.wav
+D21_823 data_thchs30/test/D21_823.wav
+D21_824 data_thchs30/test/D21_824.wav
+D21_825 data_thchs30/test/D21_825.wav
+D21_826 data_thchs30/test/D21_826.wav
+D21_827 data_thchs30/test/D21_827.wav
+D21_828 data_thchs30/test/D21_828.wav
+D21_829 data_thchs30/test/D21_829.wav
+D21_830 data_thchs30/test/D21_830.wav
+D21_831 data_thchs30/test/D21_831.wav
+D21_832 data_thchs30/test/D21_832.wav
+D21_833 data_thchs30/test/D21_833.wav
+D21_834 data_thchs30/test/D21_834.wav
+D21_835 data_thchs30/test/D21_835.wav
+D21_836 data_thchs30/test/D21_836.wav
+D21_837 data_thchs30/test/D21_837.wav
+D21_838 data_thchs30/test/D21_838.wav
+D21_839 data_thchs30/test/D21_839.wav
+D21_840 data_thchs30/test/D21_840.wav
+D21_841 data_thchs30/test/D21_841.wav
+D21_842 data_thchs30/test/D21_842.wav
+D21_843 data_thchs30/test/D21_843.wav
+D21_844 data_thchs30/test/D21_844.wav
+D21_845 data_thchs30/test/D21_845.wav
+D21_846 data_thchs30/test/D21_846.wav
+D21_847 data_thchs30/test/D21_847.wav
+D21_848 data_thchs30/test/D21_848.wav
+D21_849 data_thchs30/test/D21_849.wav
+D21_850 data_thchs30/test/D21_850.wav
+D21_851 data_thchs30/test/D21_851.wav
+D21_852 data_thchs30/test/D21_852.wav
+D21_853 data_thchs30/test/D21_853.wav
+D21_854 data_thchs30/test/D21_854.wav
+D21_855 data_thchs30/test/D21_855.wav
+D21_856 data_thchs30/test/D21_856.wav
+D21_857 data_thchs30/test/D21_857.wav
+D21_858 data_thchs30/test/D21_858.wav
+D21_859 data_thchs30/test/D21_859.wav
+D21_860 data_thchs30/test/D21_860.wav
+D21_861 data_thchs30/test/D21_861.wav
+D21_862 data_thchs30/test/D21_862.wav
+D21_863 data_thchs30/test/D21_863.wav
+D21_864 data_thchs30/test/D21_864.wav
+D21_865 data_thchs30/test/D21_865.wav
+D21_866 data_thchs30/test/D21_866.wav
+D21_867 data_thchs30/test/D21_867.wav
+D21_868 data_thchs30/test/D21_868.wav
+D21_869 data_thchs30/test/D21_869.wav
+D21_870 data_thchs30/test/D21_870.wav
+D21_871 data_thchs30/test/D21_871.wav
+D21_872 data_thchs30/test/D21_872.wav
+D21_873 data_thchs30/test/D21_873.wav
+D21_874 data_thchs30/test/D21_874.wav
+D21_875 data_thchs30/test/D21_875.wav
+D21_876 data_thchs30/test/D21_876.wav
+D21_877 data_thchs30/test/D21_877.wav
+D21_878 data_thchs30/test/D21_878.wav
+D21_879 data_thchs30/test/D21_879.wav
+D21_880 data_thchs30/test/D21_880.wav
+D21_881 data_thchs30/test/D21_881.wav
+D21_882 data_thchs30/test/D21_882.wav
+D21_883 data_thchs30/test/D21_883.wav
+D21_884 data_thchs30/test/D21_884.wav
+D21_885 data_thchs30/test/D21_885.wav
+D21_886 data_thchs30/test/D21_886.wav
+D21_887 data_thchs30/test/D21_887.wav
+D21_888 data_thchs30/test/D21_888.wav
+D21_889 data_thchs30/test/D21_889.wav
+D21_890 data_thchs30/test/D21_890.wav
+D21_891 data_thchs30/test/D21_891.wav
+D21_892 data_thchs30/test/D21_892.wav
+D21_893 data_thchs30/test/D21_893.wav
+D21_894 data_thchs30/test/D21_894.wav
+D21_895 data_thchs30/test/D21_895.wav
+D21_896 data_thchs30/test/D21_896.wav
+D21_897 data_thchs30/test/D21_897.wav
+D21_898 data_thchs30/test/D21_898.wav
+D21_899 data_thchs30/test/D21_899.wav
+D21_900 data_thchs30/test/D21_900.wav
+D21_901 data_thchs30/test/D21_901.wav
+D21_902 data_thchs30/test/D21_902.wav
+D21_903 data_thchs30/test/D21_903.wav
+D21_904 data_thchs30/test/D21_904.wav
+D21_905 data_thchs30/test/D21_905.wav
+D21_906 data_thchs30/test/D21_906.wav
+D21_907 data_thchs30/test/D21_907.wav
+D21_908 data_thchs30/test/D21_908.wav
+D21_909 data_thchs30/test/D21_909.wav
+D21_910 data_thchs30/test/D21_910.wav
+D21_911 data_thchs30/test/D21_911.wav
+D21_912 data_thchs30/test/D21_912.wav
+D21_913 data_thchs30/test/D21_913.wav
+D21_914 data_thchs30/test/D21_914.wav
+D21_915 data_thchs30/test/D21_915.wav
+D21_916 data_thchs30/test/D21_916.wav
+D21_917 data_thchs30/test/D21_917.wav
+D21_918 data_thchs30/test/D21_918.wav
+D21_919 data_thchs30/test/D21_919.wav
+D21_920 data_thchs30/test/D21_920.wav
+D21_921 data_thchs30/test/D21_921.wav
+D21_922 data_thchs30/test/D21_922.wav
+D21_923 data_thchs30/test/D21_923.wav
+D21_924 data_thchs30/test/D21_924.wav
+D21_925 data_thchs30/test/D21_925.wav
+D21_926 data_thchs30/test/D21_926.wav
+D21_927 data_thchs30/test/D21_927.wav
+D21_928 data_thchs30/test/D21_928.wav
+D21_929 data_thchs30/test/D21_929.wav
+D21_930 data_thchs30/test/D21_930.wav
+D21_931 data_thchs30/test/D21_931.wav
+D21_932 data_thchs30/test/D21_932.wav
+D21_933 data_thchs30/test/D21_933.wav
+D21_934 data_thchs30/test/D21_934.wav
+D21_935 data_thchs30/test/D21_935.wav
+D21_936 data_thchs30/test/D21_936.wav
+D21_937 data_thchs30/test/D21_937.wav
+D21_938 data_thchs30/test/D21_938.wav
+D21_939 data_thchs30/test/D21_939.wav
+D21_940 data_thchs30/test/D21_940.wav
+D21_941 data_thchs30/test/D21_941.wav
+D21_942 data_thchs30/test/D21_942.wav
+D21_943 data_thchs30/test/D21_943.wav
+D21_944 data_thchs30/test/D21_944.wav
+D21_945 data_thchs30/test/D21_945.wav
+D21_946 data_thchs30/test/D21_946.wav
+D21_947 data_thchs30/test/D21_947.wav
+D21_948 data_thchs30/test/D21_948.wav
+D21_949 data_thchs30/test/D21_949.wav
+D21_950 data_thchs30/test/D21_950.wav
+D21_951 data_thchs30/test/D21_951.wav
+D21_952 data_thchs30/test/D21_952.wav
+D21_953 data_thchs30/test/D21_953.wav
+D21_954 data_thchs30/test/D21_954.wav
+D21_955 data_thchs30/test/D21_955.wav
+D21_956 data_thchs30/test/D21_956.wav
+D21_957 data_thchs30/test/D21_957.wav
+D21_958 data_thchs30/test/D21_958.wav
+D21_959 data_thchs30/test/D21_959.wav
+D21_960 data_thchs30/test/D21_960.wav
+D21_961 data_thchs30/test/D21_961.wav
+D21_962 data_thchs30/test/D21_962.wav
+D21_963 data_thchs30/test/D21_963.wav
+D21_964 data_thchs30/test/D21_964.wav
+D21_965 data_thchs30/test/D21_965.wav
+D21_966 data_thchs30/test/D21_966.wav
+D21_967 data_thchs30/test/D21_967.wav
+D21_968 data_thchs30/test/D21_968.wav
+D21_969 data_thchs30/test/D21_969.wav
+D21_970 data_thchs30/test/D21_970.wav
+D21_971 data_thchs30/test/D21_971.wav
+D21_972 data_thchs30/test/D21_972.wav
+D21_973 data_thchs30/test/D21_973.wav
+D21_974 data_thchs30/test/D21_974.wav
+D21_975 data_thchs30/test/D21_975.wav
+D21_976 data_thchs30/test/D21_976.wav
+D21_977 data_thchs30/test/D21_977.wav
+D21_978 data_thchs30/test/D21_978.wav
+D21_979 data_thchs30/test/D21_979.wav
+D21_980 data_thchs30/test/D21_980.wav
+D21_981 data_thchs30/test/D21_981.wav
+D21_982 data_thchs30/test/D21_982.wav
+D21_983 data_thchs30/test/D21_983.wav
+D21_984 data_thchs30/test/D21_984.wav
+D21_985 data_thchs30/test/D21_985.wav
+D21_986 data_thchs30/test/D21_986.wav
+D21_987 data_thchs30/test/D21_987.wav
+D21_988 data_thchs30/test/D21_988.wav
+D21_989 data_thchs30/test/D21_989.wav
+D21_990 data_thchs30/test/D21_990.wav
+D21_991 data_thchs30/test/D21_991.wav
+D21_992 data_thchs30/test/D21_992.wav
+D21_993 data_thchs30/test/D21_993.wav
+D21_994 data_thchs30/test/D21_994.wav
+D21_995 data_thchs30/test/D21_995.wav
+D21_996 data_thchs30/test/D21_996.wav
+D21_997 data_thchs30/test/D21_997.wav
+D21_998 data_thchs30/test/D21_998.wav
+D21_999 data_thchs30/test/D21_999.wav
+D31_750 data_thchs30/test/D31_750.wav
+D31_751 data_thchs30/test/D31_751.wav
+D31_752 data_thchs30/test/D31_752.wav
+D31_753 data_thchs30/test/D31_753.wav
+D31_754 data_thchs30/test/D31_754.wav
+D31_755 data_thchs30/test/D31_755.wav
+D31_756 data_thchs30/test/D31_756.wav
+D31_757 data_thchs30/test/D31_757.wav
+D31_758 data_thchs30/test/D31_758.wav
+D31_759 data_thchs30/test/D31_759.wav
+D31_760 data_thchs30/test/D31_760.wav
+D31_761 data_thchs30/test/D31_761.wav
+D31_762 data_thchs30/test/D31_762.wav
+D31_763 data_thchs30/test/D31_763.wav
+D31_764 data_thchs30/test/D31_764.wav
+D31_765 data_thchs30/test/D31_765.wav
+D31_766 data_thchs30/test/D31_766.wav
+D31_767 data_thchs30/test/D31_767.wav
+D31_768 data_thchs30/test/D31_768.wav
+D31_769 data_thchs30/test/D31_769.wav
+D31_770 data_thchs30/test/D31_770.wav
+D31_771 data_thchs30/test/D31_771.wav
+D31_772 data_thchs30/test/D31_772.wav
+D31_773 data_thchs30/test/D31_773.wav
+D31_774 data_thchs30/test/D31_774.wav
+D31_775 data_thchs30/test/D31_775.wav
+D31_776 data_thchs30/test/D31_776.wav
+D31_777 data_thchs30/test/D31_777.wav
+D31_778 data_thchs30/test/D31_778.wav
+D31_779 data_thchs30/test/D31_779.wav
+D31_780 data_thchs30/test/D31_780.wav
+D31_781 data_thchs30/test/D31_781.wav
+D31_782 data_thchs30/test/D31_782.wav
+D31_783 data_thchs30/test/D31_783.wav
+D31_784 data_thchs30/test/D31_784.wav
+D31_785 data_thchs30/test/D31_785.wav
+D31_786 data_thchs30/test/D31_786.wav
+D31_787 data_thchs30/test/D31_787.wav
+D31_788 data_thchs30/test/D31_788.wav
+D31_789 data_thchs30/test/D31_789.wav
+D31_790 data_thchs30/test/D31_790.wav
+D31_791 data_thchs30/test/D31_791.wav
+D31_792 data_thchs30/test/D31_792.wav
+D31_793 data_thchs30/test/D31_793.wav
+D31_794 data_thchs30/test/D31_794.wav
+D31_795 data_thchs30/test/D31_795.wav
+D31_796 data_thchs30/test/D31_796.wav
+D31_797 data_thchs30/test/D31_797.wav
+D31_798 data_thchs30/test/D31_798.wav
+D31_799 data_thchs30/test/D31_799.wav
+D31_800 data_thchs30/test/D31_800.wav
+D31_801 data_thchs30/test/D31_801.wav
+D31_802 data_thchs30/test/D31_802.wav
+D31_803 data_thchs30/test/D31_803.wav
+D31_804 data_thchs30/test/D31_804.wav
+D31_805 data_thchs30/test/D31_805.wav
+D31_806 data_thchs30/test/D31_806.wav
+D31_807 data_thchs30/test/D31_807.wav
+D31_808 data_thchs30/test/D31_808.wav
+D31_809 data_thchs30/test/D31_809.wav
+D31_810 data_thchs30/test/D31_810.wav
+D31_811 data_thchs30/test/D31_811.wav
+D31_812 data_thchs30/test/D31_812.wav
+D31_813 data_thchs30/test/D31_813.wav
+D31_814 data_thchs30/test/D31_814.wav
+D31_815 data_thchs30/test/D31_815.wav
+D31_816 data_thchs30/test/D31_816.wav
+D31_817 data_thchs30/test/D31_817.wav
+D31_818 data_thchs30/test/D31_818.wav
+D31_819 data_thchs30/test/D31_819.wav
+D31_820 data_thchs30/test/D31_820.wav
+D31_821 data_thchs30/test/D31_821.wav
+D31_822 data_thchs30/test/D31_822.wav
+D31_823 data_thchs30/test/D31_823.wav
+D31_824 data_thchs30/test/D31_824.wav
+D31_825 data_thchs30/test/D31_825.wav
+D31_826 data_thchs30/test/D31_826.wav
+D31_827 data_thchs30/test/D31_827.wav
+D31_828 data_thchs30/test/D31_828.wav
+D31_829 data_thchs30/test/D31_829.wav
+D31_830 data_thchs30/test/D31_830.wav
+D31_831 data_thchs30/test/D31_831.wav
+D31_832 data_thchs30/test/D31_832.wav
+D31_833 data_thchs30/test/D31_833.wav
+D31_834 data_thchs30/test/D31_834.wav
+D31_835 data_thchs30/test/D31_835.wav
+D31_836 data_thchs30/test/D31_836.wav
+D31_837 data_thchs30/test/D31_837.wav
+D31_838 data_thchs30/test/D31_838.wav
+D31_839 data_thchs30/test/D31_839.wav
+D31_840 data_thchs30/test/D31_840.wav
+D31_841 data_thchs30/test/D31_841.wav
+D31_842 data_thchs30/test/D31_842.wav
+D31_843 data_thchs30/test/D31_843.wav
+D31_844 data_thchs30/test/D31_844.wav
+D31_845 data_thchs30/test/D31_845.wav
+D31_846 data_thchs30/test/D31_846.wav
+D31_847 data_thchs30/test/D31_847.wav
+D31_848 data_thchs30/test/D31_848.wav
+D31_849 data_thchs30/test/D31_849.wav
+D31_850 data_thchs30/test/D31_850.wav
+D31_851 data_thchs30/test/D31_851.wav
+D31_852 data_thchs30/test/D31_852.wav
+D31_853 data_thchs30/test/D31_853.wav
+D31_854 data_thchs30/test/D31_854.wav
+D31_855 data_thchs30/test/D31_855.wav
+D31_856 data_thchs30/test/D31_856.wav
+D31_857 data_thchs30/test/D31_857.wav
+D31_858 data_thchs30/test/D31_858.wav
+D31_859 data_thchs30/test/D31_859.wav
+D31_860 data_thchs30/test/D31_860.wav
+D31_861 data_thchs30/test/D31_861.wav
+D31_862 data_thchs30/test/D31_862.wav
+D31_863 data_thchs30/test/D31_863.wav
+D31_864 data_thchs30/test/D31_864.wav
+D31_865 data_thchs30/test/D31_865.wav
+D31_866 data_thchs30/test/D31_866.wav
+D31_867 data_thchs30/test/D31_867.wav
+D31_868 data_thchs30/test/D31_868.wav
+D31_869 data_thchs30/test/D31_869.wav
+D31_870 data_thchs30/test/D31_870.wav
+D31_871 data_thchs30/test/D31_871.wav
+D31_872 data_thchs30/test/D31_872.wav
+D31_873 data_thchs30/test/D31_873.wav
+D31_874 data_thchs30/test/D31_874.wav
+D31_875 data_thchs30/test/D31_875.wav
+D31_876 data_thchs30/test/D31_876.wav
+D31_877 data_thchs30/test/D31_877.wav
+D31_878 data_thchs30/test/D31_878.wav
+D31_879 data_thchs30/test/D31_879.wav
+D31_880 data_thchs30/test/D31_880.wav
+D31_881 data_thchs30/test/D31_881.wav
+D31_882 data_thchs30/test/D31_882.wav
+D31_883 data_thchs30/test/D31_883.wav
+D31_884 data_thchs30/test/D31_884.wav
+D31_885 data_thchs30/test/D31_885.wav
+D31_886 data_thchs30/test/D31_886.wav
+D31_887 data_thchs30/test/D31_887.wav
+D31_888 data_thchs30/test/D31_888.wav
+D31_889 data_thchs30/test/D31_889.wav
+D31_890 data_thchs30/test/D31_890.wav
+D31_891 data_thchs30/test/D31_891.wav
+D31_892 data_thchs30/test/D31_892.wav
+D31_893 data_thchs30/test/D31_893.wav
+D31_894 data_thchs30/test/D31_894.wav
+D31_895 data_thchs30/test/D31_895.wav
+D31_896 data_thchs30/test/D31_896.wav
+D31_897 data_thchs30/test/D31_897.wav
+D31_898 data_thchs30/test/D31_898.wav
+D31_899 data_thchs30/test/D31_899.wav
+D31_900 data_thchs30/test/D31_900.wav
+D31_901 data_thchs30/test/D31_901.wav
+D31_902 data_thchs30/test/D31_902.wav
+D31_903 data_thchs30/test/D31_903.wav
+D31_904 data_thchs30/test/D31_904.wav
+D31_905 data_thchs30/test/D31_905.wav
+D31_906 data_thchs30/test/D31_906.wav
+D31_907 data_thchs30/test/D31_907.wav
+D31_908 data_thchs30/test/D31_908.wav
+D31_909 data_thchs30/test/D31_909.wav
+D31_910 data_thchs30/test/D31_910.wav
+D31_911 data_thchs30/test/D31_911.wav
+D31_912 data_thchs30/test/D31_912.wav
+D31_913 data_thchs30/test/D31_913.wav
+D31_914 data_thchs30/test/D31_914.wav
+D31_915 data_thchs30/test/D31_915.wav
+D31_916 data_thchs30/test/D31_916.wav
+D31_917 data_thchs30/test/D31_917.wav
+D31_918 data_thchs30/test/D31_918.wav
+D31_919 data_thchs30/test/D31_919.wav
+D31_920 data_thchs30/test/D31_920.wav
+D31_921 data_thchs30/test/D31_921.wav
+D31_922 data_thchs30/test/D31_922.wav
+D31_923 data_thchs30/test/D31_923.wav
+D31_924 data_thchs30/test/D31_924.wav
+D31_925 data_thchs30/test/D31_925.wav
+D31_926 data_thchs30/test/D31_926.wav
+D31_927 data_thchs30/test/D31_927.wav
+D31_928 data_thchs30/test/D31_928.wav
+D31_929 data_thchs30/test/D31_929.wav
+D31_930 data_thchs30/test/D31_930.wav
+D31_931 data_thchs30/test/D31_931.wav
+D31_932 data_thchs30/test/D31_932.wav
+D31_933 data_thchs30/test/D31_933.wav
+D31_934 data_thchs30/test/D31_934.wav
+D31_935 data_thchs30/test/D31_935.wav
+D31_936 data_thchs30/test/D31_936.wav
+D31_937 data_thchs30/test/D31_937.wav
+D31_938 data_thchs30/test/D31_938.wav
+D31_939 data_thchs30/test/D31_939.wav
+D31_940 data_thchs30/test/D31_940.wav
+D31_941 data_thchs30/test/D31_941.wav
+D31_942 data_thchs30/test/D31_942.wav
+D31_943 data_thchs30/test/D31_943.wav
+D31_944 data_thchs30/test/D31_944.wav
+D31_945 data_thchs30/test/D31_945.wav
+D31_946 data_thchs30/test/D31_946.wav
+D31_947 data_thchs30/test/D31_947.wav
+D31_948 data_thchs30/test/D31_948.wav
+D31_949 data_thchs30/test/D31_949.wav
+D31_950 data_thchs30/test/D31_950.wav
+D31_951 data_thchs30/test/D31_951.wav
+D31_952 data_thchs30/test/D31_952.wav
+D31_953 data_thchs30/test/D31_953.wav
+D31_954 data_thchs30/test/D31_954.wav
+D31_955 data_thchs30/test/D31_955.wav
+D31_956 data_thchs30/test/D31_956.wav
+D31_957 data_thchs30/test/D31_957.wav
+D31_958 data_thchs30/test/D31_958.wav
+D31_959 data_thchs30/test/D31_959.wav
+D31_960 data_thchs30/test/D31_960.wav
+D31_961 data_thchs30/test/D31_961.wav
+D31_962 data_thchs30/test/D31_962.wav
+D31_963 data_thchs30/test/D31_963.wav
+D31_964 data_thchs30/test/D31_964.wav
+D31_965 data_thchs30/test/D31_965.wav
+D31_966 data_thchs30/test/D31_966.wav
+D31_967 data_thchs30/test/D31_967.wav
+D31_968 data_thchs30/test/D31_968.wav
+D31_969 data_thchs30/test/D31_969.wav
+D31_970 data_thchs30/test/D31_970.wav
+D31_971 data_thchs30/test/D31_971.wav
+D31_972 data_thchs30/test/D31_972.wav
+D31_973 data_thchs30/test/D31_973.wav
+D31_974 data_thchs30/test/D31_974.wav
+D31_975 data_thchs30/test/D31_975.wav
+D31_976 data_thchs30/test/D31_976.wav
+D31_977 data_thchs30/test/D31_977.wav
+D31_978 data_thchs30/test/D31_978.wav
+D31_979 data_thchs30/test/D31_979.wav
+D31_980 data_thchs30/test/D31_980.wav
+D31_981 data_thchs30/test/D31_981.wav
+D31_982 data_thchs30/test/D31_982.wav
+D31_983 data_thchs30/test/D31_983.wav
+D31_984 data_thchs30/test/D31_984.wav
+D31_985 data_thchs30/test/D31_985.wav
+D31_986 data_thchs30/test/D31_986.wav
+D31_987 data_thchs30/test/D31_987.wav
+D31_988 data_thchs30/test/D31_988.wav
+D31_989 data_thchs30/test/D31_989.wav
+D31_990 data_thchs30/test/D31_990.wav
+D31_991 data_thchs30/test/D31_991.wav
+D31_992 data_thchs30/test/D31_992.wav
+D31_993 data_thchs30/test/D31_993.wav
+D31_994 data_thchs30/test/D31_994.wav
+D31_995 data_thchs30/test/D31_995.wav
+D31_996 data_thchs30/test/D31_996.wav
+D31_997 data_thchs30/test/D31_997.wav
+D31_998 data_thchs30/test/D31_998.wav
+D31_999 data_thchs30/test/D31_999.wav
+D32_750 data_thchs30/test/D32_750.wav
+D32_751 data_thchs30/test/D32_751.wav
+D32_752 data_thchs30/test/D32_752.wav
+D32_753 data_thchs30/test/D32_753.wav
+D32_754 data_thchs30/test/D32_754.wav
+D32_755 data_thchs30/test/D32_755.wav
+D32_756 data_thchs30/test/D32_756.wav
+D32_757 data_thchs30/test/D32_757.wav
+D32_758 data_thchs30/test/D32_758.wav
+D32_759 data_thchs30/test/D32_759.wav
+D32_760 data_thchs30/test/D32_760.wav
+D32_761 data_thchs30/test/D32_761.wav
+D32_762 data_thchs30/test/D32_762.wav
+D32_763 data_thchs30/test/D32_763.wav
+D32_764 data_thchs30/test/D32_764.wav
+D32_765 data_thchs30/test/D32_765.wav
+D32_766 data_thchs30/test/D32_766.wav
+D32_767 data_thchs30/test/D32_767.wav
+D32_768 data_thchs30/test/D32_768.wav
+D32_769 data_thchs30/test/D32_769.wav
+D32_770 data_thchs30/test/D32_770.wav
+D32_771 data_thchs30/test/D32_771.wav
+D32_772 data_thchs30/test/D32_772.wav
+D32_773 data_thchs30/test/D32_773.wav
+D32_774 data_thchs30/test/D32_774.wav
+D32_775 data_thchs30/test/D32_775.wav
+D32_776 data_thchs30/test/D32_776.wav
+D32_777 data_thchs30/test/D32_777.wav
+D32_778 data_thchs30/test/D32_778.wav
+D32_779 data_thchs30/test/D32_779.wav
+D32_780 data_thchs30/test/D32_780.wav
+D32_781 data_thchs30/test/D32_781.wav
+D32_782 data_thchs30/test/D32_782.wav
+D32_783 data_thchs30/test/D32_783.wav
+D32_784 data_thchs30/test/D32_784.wav
+D32_785 data_thchs30/test/D32_785.wav
+D32_786 data_thchs30/test/D32_786.wav
+D32_787 data_thchs30/test/D32_787.wav
+D32_788 data_thchs30/test/D32_788.wav
+D32_789 data_thchs30/test/D32_789.wav
+D32_790 data_thchs30/test/D32_790.wav
+D32_791 data_thchs30/test/D32_791.wav
+D32_792 data_thchs30/test/D32_792.wav
+D32_793 data_thchs30/test/D32_793.wav
+D32_794 data_thchs30/test/D32_794.wav
+D32_795 data_thchs30/test/D32_795.wav
+D32_796 data_thchs30/test/D32_796.wav
+D32_797 data_thchs30/test/D32_797.wav
+D32_798 data_thchs30/test/D32_798.wav
+D32_799 data_thchs30/test/D32_799.wav
+D32_800 data_thchs30/test/D32_800.wav
+D32_801 data_thchs30/test/D32_801.wav
+D32_802 data_thchs30/test/D32_802.wav
+D32_803 data_thchs30/test/D32_803.wav
+D32_804 data_thchs30/test/D32_804.wav
+D32_805 data_thchs30/test/D32_805.wav
+D32_806 data_thchs30/test/D32_806.wav
+D32_807 data_thchs30/test/D32_807.wav
+D32_808 data_thchs30/test/D32_808.wav
+D32_809 data_thchs30/test/D32_809.wav
+D32_810 data_thchs30/test/D32_810.wav
+D32_811 data_thchs30/test/D32_811.wav
+D32_812 data_thchs30/test/D32_812.wav
+D32_813 data_thchs30/test/D32_813.wav
+D32_814 data_thchs30/test/D32_814.wav
+D32_815 data_thchs30/test/D32_815.wav
+D32_816 data_thchs30/test/D32_816.wav
+D32_817 data_thchs30/test/D32_817.wav
+D32_818 data_thchs30/test/D32_818.wav
+D32_819 data_thchs30/test/D32_819.wav
+D32_820 data_thchs30/test/D32_820.wav
+D32_821 data_thchs30/test/D32_821.wav
+D32_822 data_thchs30/test/D32_822.wav
+D32_823 data_thchs30/test/D32_823.wav
+D32_824 data_thchs30/test/D32_824.wav
+D32_825 data_thchs30/test/D32_825.wav
+D32_826 data_thchs30/test/D32_826.wav
+D32_827 data_thchs30/test/D32_827.wav
+D32_828 data_thchs30/test/D32_828.wav
+D32_829 data_thchs30/test/D32_829.wav
+D32_830 data_thchs30/test/D32_830.wav
+D32_831 data_thchs30/test/D32_831.wav
+D32_832 data_thchs30/test/D32_832.wav
+D32_833 data_thchs30/test/D32_833.wav
+D32_834 data_thchs30/test/D32_834.wav
+D32_835 data_thchs30/test/D32_835.wav
+D32_836 data_thchs30/test/D32_836.wav
+D32_837 data_thchs30/test/D32_837.wav
+D32_838 data_thchs30/test/D32_838.wav
+D32_839 data_thchs30/test/D32_839.wav
+D32_840 data_thchs30/test/D32_840.wav
+D32_841 data_thchs30/test/D32_841.wav
+D32_842 data_thchs30/test/D32_842.wav
+D32_843 data_thchs30/test/D32_843.wav
+D32_844 data_thchs30/test/D32_844.wav
+D32_845 data_thchs30/test/D32_845.wav
+D32_846 data_thchs30/test/D32_846.wav
+D32_847 data_thchs30/test/D32_847.wav
+D32_848 data_thchs30/test/D32_848.wav
+D32_849 data_thchs30/test/D32_849.wav
+D32_850 data_thchs30/test/D32_850.wav
+D32_851 data_thchs30/test/D32_851.wav
+D32_852 data_thchs30/test/D32_852.wav
+D32_853 data_thchs30/test/D32_853.wav
+D32_854 data_thchs30/test/D32_854.wav
+D32_855 data_thchs30/test/D32_855.wav
+D32_856 data_thchs30/test/D32_856.wav
+D32_857 data_thchs30/test/D32_857.wav
+D32_858 data_thchs30/test/D32_858.wav
+D32_859 data_thchs30/test/D32_859.wav
+D32_860 data_thchs30/test/D32_860.wav
+D32_861 data_thchs30/test/D32_861.wav
+D32_862 data_thchs30/test/D32_862.wav
+D32_863 data_thchs30/test/D32_863.wav
+D32_864 data_thchs30/test/D32_864.wav
+D32_865 data_thchs30/test/D32_865.wav
+D32_866 data_thchs30/test/D32_866.wav
+D32_867 data_thchs30/test/D32_867.wav
+D32_868 data_thchs30/test/D32_868.wav
+D32_869 data_thchs30/test/D32_869.wav
+D32_870 data_thchs30/test/D32_870.wav
+D32_871 data_thchs30/test/D32_871.wav
+D32_872 data_thchs30/test/D32_872.wav
+D32_873 data_thchs30/test/D32_873.wav
+D32_874 data_thchs30/test/D32_874.wav
+D32_875 data_thchs30/test/D32_875.wav
+D32_876 data_thchs30/test/D32_876.wav
+D32_877 data_thchs30/test/D32_877.wav
+D32_878 data_thchs30/test/D32_878.wav
+D32_879 data_thchs30/test/D32_879.wav
+D32_880 data_thchs30/test/D32_880.wav
+D32_881 data_thchs30/test/D32_881.wav
+D32_882 data_thchs30/test/D32_882.wav
+D32_883 data_thchs30/test/D32_883.wav
+D32_884 data_thchs30/test/D32_884.wav
+D32_885 data_thchs30/test/D32_885.wav
+D32_886 data_thchs30/test/D32_886.wav
+D32_887 data_thchs30/test/D32_887.wav
+D32_888 data_thchs30/test/D32_888.wav
+D32_889 data_thchs30/test/D32_889.wav
+D32_890 data_thchs30/test/D32_890.wav
+D32_891 data_thchs30/test/D32_891.wav
+D32_892 data_thchs30/test/D32_892.wav
+D32_893 data_thchs30/test/D32_893.wav
+D32_894 data_thchs30/test/D32_894.wav
+D32_895 data_thchs30/test/D32_895.wav
+D32_896 data_thchs30/test/D32_896.wav
+D32_897 data_thchs30/test/D32_897.wav
+D32_898 data_thchs30/test/D32_898.wav
+D32_899 data_thchs30/test/D32_899.wav
+D32_900 data_thchs30/test/D32_900.wav
+D32_901 data_thchs30/test/D32_901.wav
+D32_902 data_thchs30/test/D32_902.wav
+D32_903 data_thchs30/test/D32_903.wav
+D32_904 data_thchs30/test/D32_904.wav
+D32_905 data_thchs30/test/D32_905.wav
+D32_906 data_thchs30/test/D32_906.wav
+D32_907 data_thchs30/test/D32_907.wav
+D32_908 data_thchs30/test/D32_908.wav
+D32_909 data_thchs30/test/D32_909.wav
+D32_910 data_thchs30/test/D32_910.wav
+D32_911 data_thchs30/test/D32_911.wav
+D32_912 data_thchs30/test/D32_912.wav
+D32_913 data_thchs30/test/D32_913.wav
+D32_914 data_thchs30/test/D32_914.wav
+D32_915 data_thchs30/test/D32_915.wav
+D32_916 data_thchs30/test/D32_916.wav
+D32_917 data_thchs30/test/D32_917.wav
+D32_918 data_thchs30/test/D32_918.wav
+D32_919 data_thchs30/test/D32_919.wav
+D32_920 data_thchs30/test/D32_920.wav
+D32_921 data_thchs30/test/D32_921.wav
+D32_922 data_thchs30/test/D32_922.wav
+D32_923 data_thchs30/test/D32_923.wav
+D32_924 data_thchs30/test/D32_924.wav
+D32_925 data_thchs30/test/D32_925.wav
+D32_926 data_thchs30/test/D32_926.wav
+D32_927 data_thchs30/test/D32_927.wav
+D32_928 data_thchs30/test/D32_928.wav
+D32_929 data_thchs30/test/D32_929.wav
+D32_930 data_thchs30/test/D32_930.wav
+D32_931 data_thchs30/test/D32_931.wav
+D32_932 data_thchs30/test/D32_932.wav
+D32_933 data_thchs30/test/D32_933.wav
+D32_934 data_thchs30/test/D32_934.wav
+D32_935 data_thchs30/test/D32_935.wav
+D32_936 data_thchs30/test/D32_936.wav
+D32_937 data_thchs30/test/D32_937.wav
+D32_938 data_thchs30/test/D32_938.wav
+D32_939 data_thchs30/test/D32_939.wav
+D32_940 data_thchs30/test/D32_940.wav
+D32_941 data_thchs30/test/D32_941.wav
+D32_942 data_thchs30/test/D32_942.wav
+D32_943 data_thchs30/test/D32_943.wav
+D32_944 data_thchs30/test/D32_944.wav
+D32_945 data_thchs30/test/D32_945.wav
+D32_946 data_thchs30/test/D32_946.wav
+D32_947 data_thchs30/test/D32_947.wav
+D32_948 data_thchs30/test/D32_948.wav
+D32_949 data_thchs30/test/D32_949.wav
+D32_950 data_thchs30/test/D32_950.wav
+D32_951 data_thchs30/test/D32_951.wav
+D32_952 data_thchs30/test/D32_952.wav
+D32_953 data_thchs30/test/D32_953.wav
+D32_954 data_thchs30/test/D32_954.wav
+D32_955 data_thchs30/test/D32_955.wav
+D32_956 data_thchs30/test/D32_956.wav
+D32_957 data_thchs30/test/D32_957.wav
+D32_958 data_thchs30/test/D32_958.wav
+D32_959 data_thchs30/test/D32_959.wav
+D32_960 data_thchs30/test/D32_960.wav
+D32_961 data_thchs30/test/D32_961.wav
+D32_962 data_thchs30/test/D32_962.wav
+D32_963 data_thchs30/test/D32_963.wav
+D32_964 data_thchs30/test/D32_964.wav
+D32_965 data_thchs30/test/D32_965.wav
+D32_966 data_thchs30/test/D32_966.wav
+D32_967 data_thchs30/test/D32_967.wav
+D32_968 data_thchs30/test/D32_968.wav
+D32_969 data_thchs30/test/D32_969.wav
+D32_970 data_thchs30/test/D32_970.wav
+D32_971 data_thchs30/test/D32_971.wav
+D32_972 data_thchs30/test/D32_972.wav
+D32_973 data_thchs30/test/D32_973.wav
+D32_974 data_thchs30/test/D32_974.wav
+D32_975 data_thchs30/test/D32_975.wav
+D32_976 data_thchs30/test/D32_976.wav
+D32_977 data_thchs30/test/D32_977.wav
+D32_979 data_thchs30/test/D32_979.wav
+D32_980 data_thchs30/test/D32_980.wav
+D32_981 data_thchs30/test/D32_981.wav
+D32_982 data_thchs30/test/D32_982.wav
+D32_983 data_thchs30/test/D32_983.wav
+D32_984 data_thchs30/test/D32_984.wav
+D32_985 data_thchs30/test/D32_985.wav
+D32_986 data_thchs30/test/D32_986.wav
+D32_987 data_thchs30/test/D32_987.wav
+D32_988 data_thchs30/test/D32_988.wav
+D32_989 data_thchs30/test/D32_989.wav
+D32_990 data_thchs30/test/D32_990.wav
+D32_991 data_thchs30/test/D32_991.wav
+D32_992 data_thchs30/test/D32_992.wav
+D32_993 data_thchs30/test/D32_993.wav
+D32_994 data_thchs30/test/D32_994.wav
+D32_995 data_thchs30/test/D32_995.wav
+D32_996 data_thchs30/test/D32_996.wav
+D32_997 data_thchs30/test/D32_997.wav
+D32_998 data_thchs30/test/D32_998.wav
+D32_999 data_thchs30/test/D32_999.wav
+D4_750 data_thchs30/test/D4_750.wav
+D4_751 data_thchs30/test/D4_751.wav
+D4_752 data_thchs30/test/D4_752.wav
+D4_753 data_thchs30/test/D4_753.wav
+D4_754 data_thchs30/test/D4_754.wav
+D4_755 data_thchs30/test/D4_755.wav
+D4_756 data_thchs30/test/D4_756.wav
+D4_757 data_thchs30/test/D4_757.wav
+D4_758 data_thchs30/test/D4_758.wav
+D4_759 data_thchs30/test/D4_759.wav
+D4_760 data_thchs30/test/D4_760.wav
+D4_761 data_thchs30/test/D4_761.wav
+D4_762 data_thchs30/test/D4_762.wav
+D4_763 data_thchs30/test/D4_763.wav
+D4_764 data_thchs30/test/D4_764.wav
+D4_765 data_thchs30/test/D4_765.wav
+D4_766 data_thchs30/test/D4_766.wav
+D4_767 data_thchs30/test/D4_767.wav
+D4_768 data_thchs30/test/D4_768.wav
+D4_769 data_thchs30/test/D4_769.wav
+D4_770 data_thchs30/test/D4_770.wav
+D4_771 data_thchs30/test/D4_771.wav
+D4_772 data_thchs30/test/D4_772.wav
+D4_773 data_thchs30/test/D4_773.wav
+D4_774 data_thchs30/test/D4_774.wav
+D4_775 data_thchs30/test/D4_775.wav
+D4_776 data_thchs30/test/D4_776.wav
+D4_777 data_thchs30/test/D4_777.wav
+D4_778 data_thchs30/test/D4_778.wav
+D4_779 data_thchs30/test/D4_779.wav
+D4_780 data_thchs30/test/D4_780.wav
+D4_781 data_thchs30/test/D4_781.wav
+D4_782 data_thchs30/test/D4_782.wav
+D4_783 data_thchs30/test/D4_783.wav
+D4_784 data_thchs30/test/D4_784.wav
+D4_785 data_thchs30/test/D4_785.wav
+D4_786 data_thchs30/test/D4_786.wav
+D4_787 data_thchs30/test/D4_787.wav
+D4_788 data_thchs30/test/D4_788.wav
+D4_789 data_thchs30/test/D4_789.wav
+D4_790 data_thchs30/test/D4_790.wav
+D4_791 data_thchs30/test/D4_791.wav
+D4_792 data_thchs30/test/D4_792.wav
+D4_793 data_thchs30/test/D4_793.wav
+D4_794 data_thchs30/test/D4_794.wav
+D4_795 data_thchs30/test/D4_795.wav
+D4_796 data_thchs30/test/D4_796.wav
+D4_797 data_thchs30/test/D4_797.wav
+D4_798 data_thchs30/test/D4_798.wav
+D4_799 data_thchs30/test/D4_799.wav
+D4_800 data_thchs30/test/D4_800.wav
+D4_801 data_thchs30/test/D4_801.wav
+D4_802 data_thchs30/test/D4_802.wav
+D4_803 data_thchs30/test/D4_803.wav
+D4_804 data_thchs30/test/D4_804.wav
+D4_805 data_thchs30/test/D4_805.wav
+D4_806 data_thchs30/test/D4_806.wav
+D4_807 data_thchs30/test/D4_807.wav
+D4_808 data_thchs30/test/D4_808.wav
+D4_809 data_thchs30/test/D4_809.wav
+D4_810 data_thchs30/test/D4_810.wav
+D4_811 data_thchs30/test/D4_811.wav
+D4_812 data_thchs30/test/D4_812.wav
+D4_813 data_thchs30/test/D4_813.wav
+D4_814 data_thchs30/test/D4_814.wav
+D4_815 data_thchs30/test/D4_815.wav
+D4_816 data_thchs30/test/D4_816.wav
+D4_817 data_thchs30/test/D4_817.wav
+D4_818 data_thchs30/test/D4_818.wav
+D4_819 data_thchs30/test/D4_819.wav
+D4_820 data_thchs30/test/D4_820.wav
+D4_821 data_thchs30/test/D4_821.wav
+D4_822 data_thchs30/test/D4_822.wav
+D4_823 data_thchs30/test/D4_823.wav
+D4_824 data_thchs30/test/D4_824.wav
+D4_825 data_thchs30/test/D4_825.wav
+D4_826 data_thchs30/test/D4_826.wav
+D4_827 data_thchs30/test/D4_827.wav
+D4_828 data_thchs30/test/D4_828.wav
+D4_829 data_thchs30/test/D4_829.wav
+D4_830 data_thchs30/test/D4_830.wav
+D4_831 data_thchs30/test/D4_831.wav
+D4_832 data_thchs30/test/D4_832.wav
+D4_833 data_thchs30/test/D4_833.wav
+D4_834 data_thchs30/test/D4_834.wav
+D4_835 data_thchs30/test/D4_835.wav
+D4_836 data_thchs30/test/D4_836.wav
+D4_837 data_thchs30/test/D4_837.wav
+D4_838 data_thchs30/test/D4_838.wav
+D4_839 data_thchs30/test/D4_839.wav
+D4_840 data_thchs30/test/D4_840.wav
+D4_841 data_thchs30/test/D4_841.wav
+D4_842 data_thchs30/test/D4_842.wav
+D4_843 data_thchs30/test/D4_843.wav
+D4_844 data_thchs30/test/D4_844.wav
+D4_845 data_thchs30/test/D4_845.wav
+D4_846 data_thchs30/test/D4_846.wav
+D4_847 data_thchs30/test/D4_847.wav
+D4_848 data_thchs30/test/D4_848.wav
+D4_849 data_thchs30/test/D4_849.wav
+D4_850 data_thchs30/test/D4_850.wav
+D4_851 data_thchs30/test/D4_851.wav
+D4_852 data_thchs30/test/D4_852.wav
+D4_853 data_thchs30/test/D4_853.wav
+D4_854 data_thchs30/test/D4_854.wav
+D4_855 data_thchs30/test/D4_855.wav
+D4_856 data_thchs30/test/D4_856.wav
+D4_857 data_thchs30/test/D4_857.wav
+D4_858 data_thchs30/test/D4_858.wav
+D4_859 data_thchs30/test/D4_859.wav
+D4_860 data_thchs30/test/D4_860.wav
+D4_861 data_thchs30/test/D4_861.wav
+D4_862 data_thchs30/test/D4_862.wav
+D4_863 data_thchs30/test/D4_863.wav
+D4_864 data_thchs30/test/D4_864.wav
+D4_865 data_thchs30/test/D4_865.wav
+D4_866 data_thchs30/test/D4_866.wav
+D4_867 data_thchs30/test/D4_867.wav
+D4_868 data_thchs30/test/D4_868.wav
+D4_869 data_thchs30/test/D4_869.wav
+D4_870 data_thchs30/test/D4_870.wav
+D4_871 data_thchs30/test/D4_871.wav
+D4_872 data_thchs30/test/D4_872.wav
+D4_873 data_thchs30/test/D4_873.wav
+D4_874 data_thchs30/test/D4_874.wav
+D4_875 data_thchs30/test/D4_875.wav
+D4_876 data_thchs30/test/D4_876.wav
+D4_877 data_thchs30/test/D4_877.wav
+D4_878 data_thchs30/test/D4_878.wav
+D4_879 data_thchs30/test/D4_879.wav
+D4_880 data_thchs30/test/D4_880.wav
+D4_881 data_thchs30/test/D4_881.wav
+D4_882 data_thchs30/test/D4_882.wav
+D4_883 data_thchs30/test/D4_883.wav
+D4_884 data_thchs30/test/D4_884.wav
+D4_885 data_thchs30/test/D4_885.wav
+D4_886 data_thchs30/test/D4_886.wav
+D4_887 data_thchs30/test/D4_887.wav
+D4_888 data_thchs30/test/D4_888.wav
+D4_889 data_thchs30/test/D4_889.wav
+D4_890 data_thchs30/test/D4_890.wav
+D4_891 data_thchs30/test/D4_891.wav
+D4_892 data_thchs30/test/D4_892.wav
+D4_893 data_thchs30/test/D4_893.wav
+D4_894 data_thchs30/test/D4_894.wav
+D4_895 data_thchs30/test/D4_895.wav
+D4_896 data_thchs30/test/D4_896.wav
+D4_897 data_thchs30/test/D4_897.wav
+D4_898 data_thchs30/test/D4_898.wav
+D4_899 data_thchs30/test/D4_899.wav
+D4_900 data_thchs30/test/D4_900.wav
+D4_901 data_thchs30/test/D4_901.wav
+D4_902 data_thchs30/test/D4_902.wav
+D4_903 data_thchs30/test/D4_903.wav
+D4_904 data_thchs30/test/D4_904.wav
+D4_905 data_thchs30/test/D4_905.wav
+D4_906 data_thchs30/test/D4_906.wav
+D4_907 data_thchs30/test/D4_907.wav
+D4_908 data_thchs30/test/D4_908.wav
+D4_909 data_thchs30/test/D4_909.wav
+D4_910 data_thchs30/test/D4_910.wav
+D4_911 data_thchs30/test/D4_911.wav
+D4_912 data_thchs30/test/D4_912.wav
+D4_913 data_thchs30/test/D4_913.wav
+D4_914 data_thchs30/test/D4_914.wav
+D4_915 data_thchs30/test/D4_915.wav
+D4_916 data_thchs30/test/D4_916.wav
+D4_917 data_thchs30/test/D4_917.wav
+D4_918 data_thchs30/test/D4_918.wav
+D4_919 data_thchs30/test/D4_919.wav
+D4_920 data_thchs30/test/D4_920.wav
+D4_921 data_thchs30/test/D4_921.wav
+D4_922 data_thchs30/test/D4_922.wav
+D4_923 data_thchs30/test/D4_923.wav
+D4_924 data_thchs30/test/D4_924.wav
+D4_925 data_thchs30/test/D4_925.wav
+D4_926 data_thchs30/test/D4_926.wav
+D4_927 data_thchs30/test/D4_927.wav
+D4_928 data_thchs30/test/D4_928.wav
+D4_929 data_thchs30/test/D4_929.wav
+D4_930 data_thchs30/test/D4_930.wav
+D4_931 data_thchs30/test/D4_931.wav
+D4_932 data_thchs30/test/D4_932.wav
+D4_933 data_thchs30/test/D4_933.wav
+D4_934 data_thchs30/test/D4_934.wav
+D4_935 data_thchs30/test/D4_935.wav
+D4_936 data_thchs30/test/D4_936.wav
+D4_937 data_thchs30/test/D4_937.wav
+D4_938 data_thchs30/test/D4_938.wav
+D4_939 data_thchs30/test/D4_939.wav
+D4_940 data_thchs30/test/D4_940.wav
+D4_941 data_thchs30/test/D4_941.wav
+D4_942 data_thchs30/test/D4_942.wav
+D4_943 data_thchs30/test/D4_943.wav
+D4_944 data_thchs30/test/D4_944.wav
+D4_945 data_thchs30/test/D4_945.wav
+D4_946 data_thchs30/test/D4_946.wav
+D4_947 data_thchs30/test/D4_947.wav
+D4_948 data_thchs30/test/D4_948.wav
+D4_949 data_thchs30/test/D4_949.wav
+D4_950 data_thchs30/test/D4_950.wav
+D4_951 data_thchs30/test/D4_951.wav
+D4_952 data_thchs30/test/D4_952.wav
+D4_953 data_thchs30/test/D4_953.wav
+D4_954 data_thchs30/test/D4_954.wav
+D4_955 data_thchs30/test/D4_955.wav
+D4_956 data_thchs30/test/D4_956.wav
+D4_957 data_thchs30/test/D4_957.wav
+D4_958 data_thchs30/test/D4_958.wav
+D4_959 data_thchs30/test/D4_959.wav
+D4_960 data_thchs30/test/D4_960.wav
+D4_961 data_thchs30/test/D4_961.wav
+D4_962 data_thchs30/test/D4_962.wav
+D4_963 data_thchs30/test/D4_963.wav
+D4_964 data_thchs30/test/D4_964.wav
+D4_965 data_thchs30/test/D4_965.wav
+D4_966 data_thchs30/test/D4_966.wav
+D4_967 data_thchs30/test/D4_967.wav
+D4_968 data_thchs30/test/D4_968.wav
+D4_969 data_thchs30/test/D4_969.wav
+D4_970 data_thchs30/test/D4_970.wav
+D4_971 data_thchs30/test/D4_971.wav
+D4_972 data_thchs30/test/D4_972.wav
+D4_973 data_thchs30/test/D4_973.wav
+D4_974 data_thchs30/test/D4_974.wav
+D4_975 data_thchs30/test/D4_975.wav
+D4_976 data_thchs30/test/D4_976.wav
+D4_977 data_thchs30/test/D4_977.wav
+D4_978 data_thchs30/test/D4_978.wav
+D4_979 data_thchs30/test/D4_979.wav
+D4_980 data_thchs30/test/D4_980.wav
+D4_981 data_thchs30/test/D4_981.wav
+D4_982 data_thchs30/test/D4_982.wav
+D4_983 data_thchs30/test/D4_983.wav
+D4_984 data_thchs30/test/D4_984.wav
+D4_985 data_thchs30/test/D4_985.wav
+D4_986 data_thchs30/test/D4_986.wav
+D4_987 data_thchs30/test/D4_987.wav
+D4_988 data_thchs30/test/D4_988.wav
+D4_989 data_thchs30/test/D4_989.wav
+D4_990 data_thchs30/test/D4_990.wav
+D4_991 data_thchs30/test/D4_991.wav
+D4_992 data_thchs30/test/D4_992.wav
+D4_993 data_thchs30/test/D4_993.wav
+D4_994 data_thchs30/test/D4_994.wav
+D4_995 data_thchs30/test/D4_995.wav
+D4_996 data_thchs30/test/D4_996.wav
+D4_997 data_thchs30/test/D4_997.wav
+D4_998 data_thchs30/test/D4_998.wav
+D4_999 data_thchs30/test/D4_999.wav
+D6_750 data_thchs30/test/D6_750.wav
+D6_751 data_thchs30/test/D6_751.wav
+D6_752 data_thchs30/test/D6_752.wav
+D6_753 data_thchs30/test/D6_753.wav
+D6_754 data_thchs30/test/D6_754.wav
+D6_755 data_thchs30/test/D6_755.wav
+D6_756 data_thchs30/test/D6_756.wav
+D6_757 data_thchs30/test/D6_757.wav
+D6_758 data_thchs30/test/D6_758.wav
+D6_759 data_thchs30/test/D6_759.wav
+D6_760 data_thchs30/test/D6_760.wav
+D6_761 data_thchs30/test/D6_761.wav
+D6_762 data_thchs30/test/D6_762.wav
+D6_763 data_thchs30/test/D6_763.wav
+D6_764 data_thchs30/test/D6_764.wav
+D6_765 data_thchs30/test/D6_765.wav
+D6_766 data_thchs30/test/D6_766.wav
+D6_767 data_thchs30/test/D6_767.wav
+D6_768 data_thchs30/test/D6_768.wav
+D6_769 data_thchs30/test/D6_769.wav
+D6_770 data_thchs30/test/D6_770.wav
+D6_771 data_thchs30/test/D6_771.wav
+D6_772 data_thchs30/test/D6_772.wav
+D6_773 data_thchs30/test/D6_773.wav
+D6_774 data_thchs30/test/D6_774.wav
+D6_775 data_thchs30/test/D6_775.wav
+D6_776 data_thchs30/test/D6_776.wav
+D6_777 data_thchs30/test/D6_777.wav
+D6_778 data_thchs30/test/D6_778.wav
+D6_779 data_thchs30/test/D6_779.wav
+D6_780 data_thchs30/test/D6_780.wav
+D6_781 data_thchs30/test/D6_781.wav
+D6_782 data_thchs30/test/D6_782.wav
+D6_783 data_thchs30/test/D6_783.wav
+D6_784 data_thchs30/test/D6_784.wav
+D6_785 data_thchs30/test/D6_785.wav
+D6_786 data_thchs30/test/D6_786.wav
+D6_787 data_thchs30/test/D6_787.wav
+D6_788 data_thchs30/test/D6_788.wav
+D6_789 data_thchs30/test/D6_789.wav
+D6_790 data_thchs30/test/D6_790.wav
+D6_791 data_thchs30/test/D6_791.wav
+D6_792 data_thchs30/test/D6_792.wav
+D6_793 data_thchs30/test/D6_793.wav
+D6_794 data_thchs30/test/D6_794.wav
+D6_795 data_thchs30/test/D6_795.wav
+D6_796 data_thchs30/test/D6_796.wav
+D6_797 data_thchs30/test/D6_797.wav
+D6_798 data_thchs30/test/D6_798.wav
+D6_799 data_thchs30/test/D6_799.wav
+D6_800 data_thchs30/test/D6_800.wav
+D6_801 data_thchs30/test/D6_801.wav
+D6_802 data_thchs30/test/D6_802.wav
+D6_803 data_thchs30/test/D6_803.wav
+D6_804 data_thchs30/test/D6_804.wav
+D6_805 data_thchs30/test/D6_805.wav
+D6_806 data_thchs30/test/D6_806.wav
+D6_807 data_thchs30/test/D6_807.wav
+D6_808 data_thchs30/test/D6_808.wav
+D6_809 data_thchs30/test/D6_809.wav
+D6_810 data_thchs30/test/D6_810.wav
+D6_811 data_thchs30/test/D6_811.wav
+D6_812 data_thchs30/test/D6_812.wav
+D6_813 data_thchs30/test/D6_813.wav
+D6_814 data_thchs30/test/D6_814.wav
+D6_815 data_thchs30/test/D6_815.wav
+D6_816 data_thchs30/test/D6_816.wav
+D6_817 data_thchs30/test/D6_817.wav
+D6_818 data_thchs30/test/D6_818.wav
+D6_819 data_thchs30/test/D6_819.wav
+D6_820 data_thchs30/test/D6_820.wav
+D6_821 data_thchs30/test/D6_821.wav
+D6_822 data_thchs30/test/D6_822.wav
+D6_823 data_thchs30/test/D6_823.wav
+D6_824 data_thchs30/test/D6_824.wav
+D6_825 data_thchs30/test/D6_825.wav
+D6_826 data_thchs30/test/D6_826.wav
+D6_827 data_thchs30/test/D6_827.wav
+D6_828 data_thchs30/test/D6_828.wav
+D6_829 data_thchs30/test/D6_829.wav
+D6_830 data_thchs30/test/D6_830.wav
+D6_831 data_thchs30/test/D6_831.wav
+D6_832 data_thchs30/test/D6_832.wav
+D6_833 data_thchs30/test/D6_833.wav
+D6_834 data_thchs30/test/D6_834.wav
+D6_835 data_thchs30/test/D6_835.wav
+D6_836 data_thchs30/test/D6_836.wav
+D6_837 data_thchs30/test/D6_837.wav
+D6_838 data_thchs30/test/D6_838.wav
+D6_839 data_thchs30/test/D6_839.wav
+D6_840 data_thchs30/test/D6_840.wav
+D6_841 data_thchs30/test/D6_841.wav
+D6_842 data_thchs30/test/D6_842.wav
+D6_843 data_thchs30/test/D6_843.wav
+D6_844 data_thchs30/test/D6_844.wav
+D6_845 data_thchs30/test/D6_845.wav
+D6_846 data_thchs30/test/D6_846.wav
+D6_847 data_thchs30/test/D6_847.wav
+D6_848 data_thchs30/test/D6_848.wav
+D6_849 data_thchs30/test/D6_849.wav
+D6_850 data_thchs30/test/D6_850.wav
+D6_851 data_thchs30/test/D6_851.wav
+D6_852 data_thchs30/test/D6_852.wav
+D6_853 data_thchs30/test/D6_853.wav
+D6_854 data_thchs30/test/D6_854.wav
+D6_855 data_thchs30/test/D6_855.wav
+D6_856 data_thchs30/test/D6_856.wav
+D6_857 data_thchs30/test/D6_857.wav
+D6_858 data_thchs30/test/D6_858.wav
+D6_859 data_thchs30/test/D6_859.wav
+D6_860 data_thchs30/test/D6_860.wav
+D6_861 data_thchs30/test/D6_861.wav
+D6_862 data_thchs30/test/D6_862.wav
+D6_863 data_thchs30/test/D6_863.wav
+D6_864 data_thchs30/test/D6_864.wav
+D6_865 data_thchs30/test/D6_865.wav
+D6_866 data_thchs30/test/D6_866.wav
+D6_867 data_thchs30/test/D6_867.wav
+D6_868 data_thchs30/test/D6_868.wav
+D6_869 data_thchs30/test/D6_869.wav
+D6_870 data_thchs30/test/D6_870.wav
+D6_871 data_thchs30/test/D6_871.wav
+D6_872 data_thchs30/test/D6_872.wav
+D6_873 data_thchs30/test/D6_873.wav
+D6_874 data_thchs30/test/D6_874.wav
+D6_875 data_thchs30/test/D6_875.wav
+D6_876 data_thchs30/test/D6_876.wav
+D6_877 data_thchs30/test/D6_877.wav
+D6_878 data_thchs30/test/D6_878.wav
+D6_879 data_thchs30/test/D6_879.wav
+D6_880 data_thchs30/test/D6_880.wav
+D6_881 data_thchs30/test/D6_881.wav
+D6_882 data_thchs30/test/D6_882.wav
+D6_883 data_thchs30/test/D6_883.wav
+D6_884 data_thchs30/test/D6_884.wav
+D6_885 data_thchs30/test/D6_885.wav
+D6_886 data_thchs30/test/D6_886.wav
+D6_887 data_thchs30/test/D6_887.wav
+D6_888 data_thchs30/test/D6_888.wav
+D6_889 data_thchs30/test/D6_889.wav
+D6_890 data_thchs30/test/D6_890.wav
+D6_891 data_thchs30/test/D6_891.wav
+D6_892 data_thchs30/test/D6_892.wav
+D6_893 data_thchs30/test/D6_893.wav
+D6_894 data_thchs30/test/D6_894.wav
+D6_895 data_thchs30/test/D6_895.wav
+D6_896 data_thchs30/test/D6_896.wav
+D6_897 data_thchs30/test/D6_897.wav
+D6_898 data_thchs30/test/D6_898.wav
+D6_899 data_thchs30/test/D6_899.wav
+D6_900 data_thchs30/test/D6_900.wav
+D6_901 data_thchs30/test/D6_901.wav
+D6_902 data_thchs30/test/D6_902.wav
+D6_903 data_thchs30/test/D6_903.wav
+D6_904 data_thchs30/test/D6_904.wav
+D6_905 data_thchs30/test/D6_905.wav
+D6_906 data_thchs30/test/D6_906.wav
+D6_907 data_thchs30/test/D6_907.wav
+D6_908 data_thchs30/test/D6_908.wav
+D6_909 data_thchs30/test/D6_909.wav
+D6_910 data_thchs30/test/D6_910.wav
+D6_911 data_thchs30/test/D6_911.wav
+D6_912 data_thchs30/test/D6_912.wav
+D6_913 data_thchs30/test/D6_913.wav
+D6_914 data_thchs30/test/D6_914.wav
+D6_915 data_thchs30/test/D6_915.wav
+D6_916 data_thchs30/test/D6_916.wav
+D6_917 data_thchs30/test/D6_917.wav
+D6_918 data_thchs30/test/D6_918.wav
+D6_919 data_thchs30/test/D6_919.wav
+D6_920 data_thchs30/test/D6_920.wav
+D6_921 data_thchs30/test/D6_921.wav
+D6_922 data_thchs30/test/D6_922.wav
+D6_923 data_thchs30/test/D6_923.wav
+D6_924 data_thchs30/test/D6_924.wav
+D6_925 data_thchs30/test/D6_925.wav
+D6_926 data_thchs30/test/D6_926.wav
+D6_927 data_thchs30/test/D6_927.wav
+D6_928 data_thchs30/test/D6_928.wav
+D6_929 data_thchs30/test/D6_929.wav
+D6_930 data_thchs30/test/D6_930.wav
+D6_931 data_thchs30/test/D6_931.wav
+D6_932 data_thchs30/test/D6_932.wav
+D6_933 data_thchs30/test/D6_933.wav
+D6_934 data_thchs30/test/D6_934.wav
+D6_935 data_thchs30/test/D6_935.wav
+D6_936 data_thchs30/test/D6_936.wav
+D6_937 data_thchs30/test/D6_937.wav
+D6_938 data_thchs30/test/D6_938.wav
+D6_939 data_thchs30/test/D6_939.wav
+D6_940 data_thchs30/test/D6_940.wav
+D6_941 data_thchs30/test/D6_941.wav
+D6_942 data_thchs30/test/D6_942.wav
+D6_943 data_thchs30/test/D6_943.wav
+D6_944 data_thchs30/test/D6_944.wav
+D6_945 data_thchs30/test/D6_945.wav
+D6_946 data_thchs30/test/D6_946.wav
+D6_947 data_thchs30/test/D6_947.wav
+D6_948 data_thchs30/test/D6_948.wav
+D6_949 data_thchs30/test/D6_949.wav
+D6_950 data_thchs30/test/D6_950.wav
+D6_951 data_thchs30/test/D6_951.wav
+D6_952 data_thchs30/test/D6_952.wav
+D6_953 data_thchs30/test/D6_953.wav
+D6_954 data_thchs30/test/D6_954.wav
+D6_955 data_thchs30/test/D6_955.wav
+D6_956 data_thchs30/test/D6_956.wav
+D6_957 data_thchs30/test/D6_957.wav
+D6_958 data_thchs30/test/D6_958.wav
+D6_959 data_thchs30/test/D6_959.wav
+D6_960 data_thchs30/test/D6_960.wav
+D6_961 data_thchs30/test/D6_961.wav
+D6_962 data_thchs30/test/D6_962.wav
+D6_963 data_thchs30/test/D6_963.wav
+D6_964 data_thchs30/test/D6_964.wav
+D6_965 data_thchs30/test/D6_965.wav
+D6_966 data_thchs30/test/D6_966.wav
+D6_967 data_thchs30/test/D6_967.wav
+D6_968 data_thchs30/test/D6_968.wav
+D6_969 data_thchs30/test/D6_969.wav
+D6_970 data_thchs30/test/D6_970.wav
+D6_971 data_thchs30/test/D6_971.wav
+D6_972 data_thchs30/test/D6_972.wav
+D6_973 data_thchs30/test/D6_973.wav
+D6_974 data_thchs30/test/D6_974.wav
+D6_975 data_thchs30/test/D6_975.wav
+D6_976 data_thchs30/test/D6_976.wav
+D6_977 data_thchs30/test/D6_977.wav
+D6_978 data_thchs30/test/D6_978.wav
+D6_979 data_thchs30/test/D6_979.wav
+D6_980 data_thchs30/test/D6_980.wav
+D6_981 data_thchs30/test/D6_981.wav
+D6_982 data_thchs30/test/D6_982.wav
+D6_983 data_thchs30/test/D6_983.wav
+D6_984 data_thchs30/test/D6_984.wav
+D6_985 data_thchs30/test/D6_985.wav
+D6_986 data_thchs30/test/D6_986.wav
+D6_987 data_thchs30/test/D6_987.wav
+D6_988 data_thchs30/test/D6_988.wav
+D6_989 data_thchs30/test/D6_989.wav
+D6_990 data_thchs30/test/D6_990.wav
+D6_991 data_thchs30/test/D6_991.wav
+D6_992 data_thchs30/test/D6_992.wav
+D6_993 data_thchs30/test/D6_993.wav
+D6_994 data_thchs30/test/D6_994.wav
+D6_995 data_thchs30/test/D6_995.wav
+D6_996 data_thchs30/test/D6_996.wav
+D6_997 data_thchs30/test/D6_997.wav
+D6_998 data_thchs30/test/D6_998.wav
+D6_999 data_thchs30/test/D6_999.wav
+D7_750 data_thchs30/test/D7_750.wav
+D7_751 data_thchs30/test/D7_751.wav
+D7_752 data_thchs30/test/D7_752.wav
+D7_753 data_thchs30/test/D7_753.wav
+D7_754 data_thchs30/test/D7_754.wav
+D7_755 data_thchs30/test/D7_755.wav
+D7_756 data_thchs30/test/D7_756.wav
+D7_757 data_thchs30/test/D7_757.wav
+D7_758 data_thchs30/test/D7_758.wav
+D7_759 data_thchs30/test/D7_759.wav
+D7_760 data_thchs30/test/D7_760.wav
+D7_761 data_thchs30/test/D7_761.wav
+D7_762 data_thchs30/test/D7_762.wav
+D7_763 data_thchs30/test/D7_763.wav
+D7_764 data_thchs30/test/D7_764.wav
+D7_765 data_thchs30/test/D7_765.wav
+D7_766 data_thchs30/test/D7_766.wav
+D7_767 data_thchs30/test/D7_767.wav
+D7_768 data_thchs30/test/D7_768.wav
+D7_769 data_thchs30/test/D7_769.wav
+D7_770 data_thchs30/test/D7_770.wav
+D7_771 data_thchs30/test/D7_771.wav
+D7_772 data_thchs30/test/D7_772.wav
+D7_773 data_thchs30/test/D7_773.wav
+D7_774 data_thchs30/test/D7_774.wav
+D7_775 data_thchs30/test/D7_775.wav
+D7_776 data_thchs30/test/D7_776.wav
+D7_777 data_thchs30/test/D7_777.wav
+D7_778 data_thchs30/test/D7_778.wav
+D7_779 data_thchs30/test/D7_779.wav
+D7_780 data_thchs30/test/D7_780.wav
+D7_781 data_thchs30/test/D7_781.wav
+D7_782 data_thchs30/test/D7_782.wav
+D7_783 data_thchs30/test/D7_783.wav
+D7_784 data_thchs30/test/D7_784.wav
+D7_785 data_thchs30/test/D7_785.wav
+D7_786 data_thchs30/test/D7_786.wav
+D7_787 data_thchs30/test/D7_787.wav
+D7_788 data_thchs30/test/D7_788.wav
+D7_789 data_thchs30/test/D7_789.wav
+D7_790 data_thchs30/test/D7_790.wav
+D7_791 data_thchs30/test/D7_791.wav
+D7_792 data_thchs30/test/D7_792.wav
+D7_793 data_thchs30/test/D7_793.wav
+D7_794 data_thchs30/test/D7_794.wav
+D7_795 data_thchs30/test/D7_795.wav
+D7_796 data_thchs30/test/D7_796.wav
+D7_797 data_thchs30/test/D7_797.wav
+D7_798 data_thchs30/test/D7_798.wav
+D7_799 data_thchs30/test/D7_799.wav
+D7_800 data_thchs30/test/D7_800.wav
+D7_801 data_thchs30/test/D7_801.wav
+D7_802 data_thchs30/test/D7_802.wav
+D7_803 data_thchs30/test/D7_803.wav
+D7_804 data_thchs30/test/D7_804.wav
+D7_805 data_thchs30/test/D7_805.wav
+D7_806 data_thchs30/test/D7_806.wav
+D7_807 data_thchs30/test/D7_807.wav
+D7_808 data_thchs30/test/D7_808.wav
+D7_809 data_thchs30/test/D7_809.wav
+D7_810 data_thchs30/test/D7_810.wav
+D7_811 data_thchs30/test/D7_811.wav
+D7_812 data_thchs30/test/D7_812.wav
+D7_813 data_thchs30/test/D7_813.wav
+D7_814 data_thchs30/test/D7_814.wav
+D7_815 data_thchs30/test/D7_815.wav
+D7_816 data_thchs30/test/D7_816.wav
+D7_817 data_thchs30/test/D7_817.wav
+D7_818 data_thchs30/test/D7_818.wav
+D7_819 data_thchs30/test/D7_819.wav
+D7_820 data_thchs30/test/D7_820.wav
+D7_821 data_thchs30/test/D7_821.wav
+D7_822 data_thchs30/test/D7_822.wav
+D7_823 data_thchs30/test/D7_823.wav
+D7_824 data_thchs30/test/D7_824.wav
+D7_825 data_thchs30/test/D7_825.wav
+D7_826 data_thchs30/test/D7_826.wav
+D7_827 data_thchs30/test/D7_827.wav
+D7_828 data_thchs30/test/D7_828.wav
+D7_829 data_thchs30/test/D7_829.wav
+D7_830 data_thchs30/test/D7_830.wav
+D7_831 data_thchs30/test/D7_831.wav
+D7_832 data_thchs30/test/D7_832.wav
+D7_833 data_thchs30/test/D7_833.wav
+D7_834 data_thchs30/test/D7_834.wav
+D7_835 data_thchs30/test/D7_835.wav
+D7_836 data_thchs30/test/D7_836.wav
+D7_837 data_thchs30/test/D7_837.wav
+D7_838 data_thchs30/test/D7_838.wav
+D7_839 data_thchs30/test/D7_839.wav
+D7_840 data_thchs30/test/D7_840.wav
+D7_841 data_thchs30/test/D7_841.wav
+D7_842 data_thchs30/test/D7_842.wav
+D7_843 data_thchs30/test/D7_843.wav
+D7_844 data_thchs30/test/D7_844.wav
+D7_845 data_thchs30/test/D7_845.wav
+D7_846 data_thchs30/test/D7_846.wav
+D7_847 data_thchs30/test/D7_847.wav
+D7_848 data_thchs30/test/D7_848.wav
+D7_849 data_thchs30/test/D7_849.wav
+D7_850 data_thchs30/test/D7_850.wav
+D7_851 data_thchs30/test/D7_851.wav
+D7_852 data_thchs30/test/D7_852.wav
+D7_853 data_thchs30/test/D7_853.wav
+D7_854 data_thchs30/test/D7_854.wav
+D7_855 data_thchs30/test/D7_855.wav
+D7_856 data_thchs30/test/D7_856.wav
+D7_857 data_thchs30/test/D7_857.wav
+D7_858 data_thchs30/test/D7_858.wav
+D7_859 data_thchs30/test/D7_859.wav
+D7_860 data_thchs30/test/D7_860.wav
+D7_861 data_thchs30/test/D7_861.wav
+D7_862 data_thchs30/test/D7_862.wav
+D7_863 data_thchs30/test/D7_863.wav
+D7_864 data_thchs30/test/D7_864.wav
+D7_865 data_thchs30/test/D7_865.wav
+D7_866 data_thchs30/test/D7_866.wav
+D7_867 data_thchs30/test/D7_867.wav
+D7_868 data_thchs30/test/D7_868.wav
+D7_869 data_thchs30/test/D7_869.wav
+D7_870 data_thchs30/test/D7_870.wav
+D7_871 data_thchs30/test/D7_871.wav
+D7_872 data_thchs30/test/D7_872.wav
+D7_873 data_thchs30/test/D7_873.wav
+D7_874 data_thchs30/test/D7_874.wav
+D7_875 data_thchs30/test/D7_875.wav
+D7_876 data_thchs30/test/D7_876.wav
+D7_877 data_thchs30/test/D7_877.wav
+D7_878 data_thchs30/test/D7_878.wav
+D7_879 data_thchs30/test/D7_879.wav
+D7_880 data_thchs30/test/D7_880.wav
+D7_881 data_thchs30/test/D7_881.wav
+D7_882 data_thchs30/test/D7_882.wav
+D7_883 data_thchs30/test/D7_883.wav
+D7_884 data_thchs30/test/D7_884.wav
+D7_885 data_thchs30/test/D7_885.wav
+D7_886 data_thchs30/test/D7_886.wav
+D7_887 data_thchs30/test/D7_887.wav
+D7_888 data_thchs30/test/D7_888.wav
+D7_889 data_thchs30/test/D7_889.wav
+D7_890 data_thchs30/test/D7_890.wav
+D7_891 data_thchs30/test/D7_891.wav
+D7_892 data_thchs30/test/D7_892.wav
+D7_893 data_thchs30/test/D7_893.wav
+D7_894 data_thchs30/test/D7_894.wav
+D7_895 data_thchs30/test/D7_895.wav
+D7_896 data_thchs30/test/D7_896.wav
+D7_897 data_thchs30/test/D7_897.wav
+D7_898 data_thchs30/test/D7_898.wav
+D7_899 data_thchs30/test/D7_899.wav
+D7_900 data_thchs30/test/D7_900.wav
+D7_901 data_thchs30/test/D7_901.wav
+D7_902 data_thchs30/test/D7_902.wav
+D7_903 data_thchs30/test/D7_903.wav
+D7_904 data_thchs30/test/D7_904.wav
+D7_905 data_thchs30/test/D7_905.wav
+D7_906 data_thchs30/test/D7_906.wav
+D7_907 data_thchs30/test/D7_907.wav
+D7_908 data_thchs30/test/D7_908.wav
+D7_909 data_thchs30/test/D7_909.wav
+D7_910 data_thchs30/test/D7_910.wav
+D7_911 data_thchs30/test/D7_911.wav
+D7_912 data_thchs30/test/D7_912.wav
+D7_913 data_thchs30/test/D7_913.wav
+D7_914 data_thchs30/test/D7_914.wav
+D7_915 data_thchs30/test/D7_915.wav
+D7_916 data_thchs30/test/D7_916.wav
+D7_917 data_thchs30/test/D7_917.wav
+D7_918 data_thchs30/test/D7_918.wav
+D7_919 data_thchs30/test/D7_919.wav
+D7_920 data_thchs30/test/D7_920.wav
+D7_921 data_thchs30/test/D7_921.wav
+D7_922 data_thchs30/test/D7_922.wav
+D7_923 data_thchs30/test/D7_923.wav
+D7_924 data_thchs30/test/D7_924.wav
+D7_925 data_thchs30/test/D7_925.wav
+D7_926 data_thchs30/test/D7_926.wav
+D7_927 data_thchs30/test/D7_927.wav
+D7_928 data_thchs30/test/D7_928.wav
+D7_929 data_thchs30/test/D7_929.wav
+D7_930 data_thchs30/test/D7_930.wav
+D7_931 data_thchs30/test/D7_931.wav
+D7_932 data_thchs30/test/D7_932.wav
+D7_933 data_thchs30/test/D7_933.wav
+D7_934 data_thchs30/test/D7_934.wav
+D7_935 data_thchs30/test/D7_935.wav
+D7_936 data_thchs30/test/D7_936.wav
+D7_937 data_thchs30/test/D7_937.wav
+D7_938 data_thchs30/test/D7_938.wav
+D7_939 data_thchs30/test/D7_939.wav
+D7_940 data_thchs30/test/D7_940.wav
+D7_941 data_thchs30/test/D7_941.wav
+D7_942 data_thchs30/test/D7_942.wav
+D7_943 data_thchs30/test/D7_943.wav
+D7_944 data_thchs30/test/D7_944.wav
+D7_945 data_thchs30/test/D7_945.wav
+D7_946 data_thchs30/test/D7_946.wav
+D7_947 data_thchs30/test/D7_947.wav
+D7_948 data_thchs30/test/D7_948.wav
+D7_949 data_thchs30/test/D7_949.wav
+D7_950 data_thchs30/test/D7_950.wav
+D7_951 data_thchs30/test/D7_951.wav
+D7_952 data_thchs30/test/D7_952.wav
+D7_953 data_thchs30/test/D7_953.wav
+D7_954 data_thchs30/test/D7_954.wav
+D7_955 data_thchs30/test/D7_955.wav
+D7_956 data_thchs30/test/D7_956.wav
+D7_957 data_thchs30/test/D7_957.wav
+D7_958 data_thchs30/test/D7_958.wav
+D7_959 data_thchs30/test/D7_959.wav
+D7_960 data_thchs30/test/D7_960.wav
+D7_961 data_thchs30/test/D7_961.wav
+D7_962 data_thchs30/test/D7_962.wav
+D7_963 data_thchs30/test/D7_963.wav
+D7_964 data_thchs30/test/D7_964.wav
+D7_965 data_thchs30/test/D7_965.wav
+D7_966 data_thchs30/test/D7_966.wav
+D7_967 data_thchs30/test/D7_967.wav
+D7_968 data_thchs30/test/D7_968.wav
+D7_969 data_thchs30/test/D7_969.wav
+D7_970 data_thchs30/test/D7_970.wav
+D7_971 data_thchs30/test/D7_971.wav
+D7_972 data_thchs30/test/D7_972.wav
+D7_973 data_thchs30/test/D7_973.wav
+D7_974 data_thchs30/test/D7_974.wav
+D7_975 data_thchs30/test/D7_975.wav
+D7_976 data_thchs30/test/D7_976.wav
+D7_977 data_thchs30/test/D7_977.wav
+D7_978 data_thchs30/test/D7_978.wav
+D7_979 data_thchs30/test/D7_979.wav
+D7_980 data_thchs30/test/D7_980.wav
+D7_981 data_thchs30/test/D7_981.wav
+D7_982 data_thchs30/test/D7_982.wav
+D7_983 data_thchs30/test/D7_983.wav
+D7_984 data_thchs30/test/D7_984.wav
+D7_985 data_thchs30/test/D7_985.wav
+D7_986 data_thchs30/test/D7_986.wav
+D7_987 data_thchs30/test/D7_987.wav
+D7_988 data_thchs30/test/D7_988.wav
+D7_989 data_thchs30/test/D7_989.wav
+D7_990 data_thchs30/test/D7_990.wav
+D7_991 data_thchs30/test/D7_991.wav
+D7_992 data_thchs30/test/D7_992.wav
+D7_993 data_thchs30/test/D7_993.wav
+D7_994 data_thchs30/test/D7_994.wav
+D7_995 data_thchs30/test/D7_995.wav
+D7_996 data_thchs30/test/D7_996.wav
+D7_997 data_thchs30/test/D7_997.wav
+D7_998 data_thchs30/test/D7_998.wav
+D8_750 data_thchs30/test/D8_750.wav
+D8_751 data_thchs30/test/D8_751.wav
+D8_752 data_thchs30/test/D8_752.wav
+D8_753 data_thchs30/test/D8_753.wav
+D8_754 data_thchs30/test/D8_754.wav
+D8_755 data_thchs30/test/D8_755.wav
+D8_756 data_thchs30/test/D8_756.wav
+D8_757 data_thchs30/test/D8_757.wav
+D8_758 data_thchs30/test/D8_758.wav
+D8_759 data_thchs30/test/D8_759.wav
+D8_760 data_thchs30/test/D8_760.wav
+D8_761 data_thchs30/test/D8_761.wav
+D8_762 data_thchs30/test/D8_762.wav
+D8_763 data_thchs30/test/D8_763.wav
+D8_764 data_thchs30/test/D8_764.wav
+D8_765 data_thchs30/test/D8_765.wav
+D8_766 data_thchs30/test/D8_766.wav
+D8_767 data_thchs30/test/D8_767.wav
+D8_768 data_thchs30/test/D8_768.wav
+D8_769 data_thchs30/test/D8_769.wav
+D8_770 data_thchs30/test/D8_770.wav
+D8_771 data_thchs30/test/D8_771.wav
+D8_772 data_thchs30/test/D8_772.wav
+D8_773 data_thchs30/test/D8_773.wav
+D8_774 data_thchs30/test/D8_774.wav
+D8_775 data_thchs30/test/D8_775.wav
+D8_776 data_thchs30/test/D8_776.wav
+D8_777 data_thchs30/test/D8_777.wav
+D8_778 data_thchs30/test/D8_778.wav
+D8_779 data_thchs30/test/D8_779.wav
+D8_780 data_thchs30/test/D8_780.wav
+D8_781 data_thchs30/test/D8_781.wav
+D8_782 data_thchs30/test/D8_782.wav
+D8_783 data_thchs30/test/D8_783.wav
+D8_784 data_thchs30/test/D8_784.wav
+D8_785 data_thchs30/test/D8_785.wav
+D8_786 data_thchs30/test/D8_786.wav
+D8_787 data_thchs30/test/D8_787.wav
+D8_788 data_thchs30/test/D8_788.wav
+D8_789 data_thchs30/test/D8_789.wav
+D8_790 data_thchs30/test/D8_790.wav
+D8_791 data_thchs30/test/D8_791.wav
+D8_792 data_thchs30/test/D8_792.wav
+D8_793 data_thchs30/test/D8_793.wav
+D8_794 data_thchs30/test/D8_794.wav
+D8_795 data_thchs30/test/D8_795.wav
+D8_796 data_thchs30/test/D8_796.wav
+D8_797 data_thchs30/test/D8_797.wav
+D8_798 data_thchs30/test/D8_798.wav
+D8_799 data_thchs30/test/D8_799.wav
+D8_800 data_thchs30/test/D8_800.wav
+D8_801 data_thchs30/test/D8_801.wav
+D8_802 data_thchs30/test/D8_802.wav
+D8_803 data_thchs30/test/D8_803.wav
+D8_804 data_thchs30/test/D8_804.wav
+D8_805 data_thchs30/test/D8_805.wav
+D8_806 data_thchs30/test/D8_806.wav
+D8_807 data_thchs30/test/D8_807.wav
+D8_808 data_thchs30/test/D8_808.wav
+D8_809 data_thchs30/test/D8_809.wav
+D8_810 data_thchs30/test/D8_810.wav
+D8_811 data_thchs30/test/D8_811.wav
+D8_812 data_thchs30/test/D8_812.wav
+D8_813 data_thchs30/test/D8_813.wav
+D8_814 data_thchs30/test/D8_814.wav
+D8_815 data_thchs30/test/D8_815.wav
+D8_816 data_thchs30/test/D8_816.wav
+D8_817 data_thchs30/test/D8_817.wav
+D8_818 data_thchs30/test/D8_818.wav
+D8_819 data_thchs30/test/D8_819.wav
+D8_820 data_thchs30/test/D8_820.wav
+D8_821 data_thchs30/test/D8_821.wav
+D8_822 data_thchs30/test/D8_822.wav
+D8_823 data_thchs30/test/D8_823.wav
+D8_824 data_thchs30/test/D8_824.wav
+D8_825 data_thchs30/test/D8_825.wav
+D8_826 data_thchs30/test/D8_826.wav
+D8_827 data_thchs30/test/D8_827.wav
+D8_828 data_thchs30/test/D8_828.wav
+D8_829 data_thchs30/test/D8_829.wav
+D8_830 data_thchs30/test/D8_830.wav
+D8_831 data_thchs30/test/D8_831.wav
+D8_832 data_thchs30/test/D8_832.wav
+D8_833 data_thchs30/test/D8_833.wav
+D8_834 data_thchs30/test/D8_834.wav
+D8_835 data_thchs30/test/D8_835.wav
+D8_836 data_thchs30/test/D8_836.wav
+D8_837 data_thchs30/test/D8_837.wav
+D8_838 data_thchs30/test/D8_838.wav
+D8_839 data_thchs30/test/D8_839.wav
+D8_840 data_thchs30/test/D8_840.wav
+D8_841 data_thchs30/test/D8_841.wav
+D8_842 data_thchs30/test/D8_842.wav
+D8_843 data_thchs30/test/D8_843.wav
+D8_844 data_thchs30/test/D8_844.wav
+D8_845 data_thchs30/test/D8_845.wav
+D8_846 data_thchs30/test/D8_846.wav
+D8_847 data_thchs30/test/D8_847.wav
+D8_848 data_thchs30/test/D8_848.wav
+D8_849 data_thchs30/test/D8_849.wav
+D8_850 data_thchs30/test/D8_850.wav
+D8_851 data_thchs30/test/D8_851.wav
+D8_852 data_thchs30/test/D8_852.wav
+D8_853 data_thchs30/test/D8_853.wav
+D8_854 data_thchs30/test/D8_854.wav
+D8_855 data_thchs30/test/D8_855.wav
+D8_856 data_thchs30/test/D8_856.wav
+D8_857 data_thchs30/test/D8_857.wav
+D8_858 data_thchs30/test/D8_858.wav
+D8_859 data_thchs30/test/D8_859.wav
+D8_860 data_thchs30/test/D8_860.wav
+D8_861 data_thchs30/test/D8_861.wav
+D8_862 data_thchs30/test/D8_862.wav
+D8_863 data_thchs30/test/D8_863.wav
+D8_864 data_thchs30/test/D8_864.wav
+D8_865 data_thchs30/test/D8_865.wav
+D8_866 data_thchs30/test/D8_866.wav
+D8_867 data_thchs30/test/D8_867.wav
+D8_868 data_thchs30/test/D8_868.wav
+D8_869 data_thchs30/test/D8_869.wav
+D8_870 data_thchs30/test/D8_870.wav
+D8_871 data_thchs30/test/D8_871.wav
+D8_872 data_thchs30/test/D8_872.wav
+D8_873 data_thchs30/test/D8_873.wav
+D8_874 data_thchs30/test/D8_874.wav
+D8_875 data_thchs30/test/D8_875.wav
+D8_876 data_thchs30/test/D8_876.wav
+D8_877 data_thchs30/test/D8_877.wav
+D8_878 data_thchs30/test/D8_878.wav
+D8_879 data_thchs30/test/D8_879.wav
+D8_880 data_thchs30/test/D8_880.wav
+D8_881 data_thchs30/test/D8_881.wav
+D8_882 data_thchs30/test/D8_882.wav
+D8_883 data_thchs30/test/D8_883.wav
+D8_884 data_thchs30/test/D8_884.wav
+D8_885 data_thchs30/test/D8_885.wav
+D8_886 data_thchs30/test/D8_886.wav
+D8_887 data_thchs30/test/D8_887.wav
+D8_888 data_thchs30/test/D8_888.wav
+D8_889 data_thchs30/test/D8_889.wav
+D8_890 data_thchs30/test/D8_890.wav
+D8_891 data_thchs30/test/D8_891.wav
+D8_892 data_thchs30/test/D8_892.wav
+D8_893 data_thchs30/test/D8_893.wav
+D8_894 data_thchs30/test/D8_894.wav
+D8_895 data_thchs30/test/D8_895.wav
+D8_896 data_thchs30/test/D8_896.wav
+D8_897 data_thchs30/test/D8_897.wav
+D8_898 data_thchs30/test/D8_898.wav
+D8_899 data_thchs30/test/D8_899.wav
+D8_900 data_thchs30/test/D8_900.wav
+D8_901 data_thchs30/test/D8_901.wav
+D8_902 data_thchs30/test/D8_902.wav
+D8_903 data_thchs30/test/D8_903.wav
+D8_904 data_thchs30/test/D8_904.wav
+D8_905 data_thchs30/test/D8_905.wav
+D8_906 data_thchs30/test/D8_906.wav
+D8_907 data_thchs30/test/D8_907.wav
+D8_908 data_thchs30/test/D8_908.wav
+D8_909 data_thchs30/test/D8_909.wav
+D8_910 data_thchs30/test/D8_910.wav
+D8_911 data_thchs30/test/D8_911.wav
+D8_912 data_thchs30/test/D8_912.wav
+D8_913 data_thchs30/test/D8_913.wav
+D8_914 data_thchs30/test/D8_914.wav
+D8_915 data_thchs30/test/D8_915.wav
+D8_916 data_thchs30/test/D8_916.wav
+D8_917 data_thchs30/test/D8_917.wav
+D8_918 data_thchs30/test/D8_918.wav
+D8_919 data_thchs30/test/D8_919.wav
+D8_920 data_thchs30/test/D8_920.wav
+D8_921 data_thchs30/test/D8_921.wav
+D8_922 data_thchs30/test/D8_922.wav
+D8_923 data_thchs30/test/D8_923.wav
+D8_924 data_thchs30/test/D8_924.wav
+D8_925 data_thchs30/test/D8_925.wav
+D8_926 data_thchs30/test/D8_926.wav
+D8_927 data_thchs30/test/D8_927.wav
+D8_928 data_thchs30/test/D8_928.wav
+D8_929 data_thchs30/test/D8_929.wav
+D8_930 data_thchs30/test/D8_930.wav
+D8_931 data_thchs30/test/D8_931.wav
+D8_932 data_thchs30/test/D8_932.wav
+D8_933 data_thchs30/test/D8_933.wav
+D8_934 data_thchs30/test/D8_934.wav
+D8_935 data_thchs30/test/D8_935.wav
+D8_936 data_thchs30/test/D8_936.wav
+D8_937 data_thchs30/test/D8_937.wav
+D8_938 data_thchs30/test/D8_938.wav
+D8_939 data_thchs30/test/D8_939.wav
+D8_940 data_thchs30/test/D8_940.wav
+D8_941 data_thchs30/test/D8_941.wav
+D8_942 data_thchs30/test/D8_942.wav
+D8_943 data_thchs30/test/D8_943.wav
+D8_944 data_thchs30/test/D8_944.wav
+D8_945 data_thchs30/test/D8_945.wav
+D8_946 data_thchs30/test/D8_946.wav
+D8_947 data_thchs30/test/D8_947.wav
+D8_948 data_thchs30/test/D8_948.wav
+D8_949 data_thchs30/test/D8_949.wav
+D8_950 data_thchs30/test/D8_950.wav
+D8_951 data_thchs30/test/D8_951.wav
+D8_952 data_thchs30/test/D8_952.wav
+D8_953 data_thchs30/test/D8_953.wav
+D8_954 data_thchs30/test/D8_954.wav
+D8_955 data_thchs30/test/D8_955.wav
+D8_956 data_thchs30/test/D8_956.wav
+D8_957 data_thchs30/test/D8_957.wav
+D8_958 data_thchs30/test/D8_958.wav
+D8_959 data_thchs30/test/D8_959.wav
+D8_960 data_thchs30/test/D8_960.wav
+D8_961 data_thchs30/test/D8_961.wav
+D8_962 data_thchs30/test/D8_962.wav
+D8_963 data_thchs30/test/D8_963.wav
+D8_964 data_thchs30/test/D8_964.wav
+D8_965 data_thchs30/test/D8_965.wav
+D8_966 data_thchs30/test/D8_966.wav
+D8_967 data_thchs30/test/D8_967.wav
+D8_968 data_thchs30/test/D8_968.wav
+D8_969 data_thchs30/test/D8_969.wav
+D8_970 data_thchs30/test/D8_970.wav
+D8_971 data_thchs30/test/D8_971.wav
+D8_972 data_thchs30/test/D8_972.wav
+D8_973 data_thchs30/test/D8_973.wav
+D8_974 data_thchs30/test/D8_974.wav
+D8_975 data_thchs30/test/D8_975.wav
+D8_976 data_thchs30/test/D8_976.wav
+D8_977 data_thchs30/test/D8_977.wav
+D8_978 data_thchs30/test/D8_978.wav
+D8_979 data_thchs30/test/D8_979.wav
+D8_980 data_thchs30/test/D8_980.wav
+D8_981 data_thchs30/test/D8_981.wav
+D8_982 data_thchs30/test/D8_982.wav
+D8_983 data_thchs30/test/D8_983.wav
+D8_984 data_thchs30/test/D8_984.wav
+D8_985 data_thchs30/test/D8_985.wav
+D8_986 data_thchs30/test/D8_986.wav
+D8_987 data_thchs30/test/D8_987.wav
+D8_988 data_thchs30/test/D8_988.wav
+D8_989 data_thchs30/test/D8_989.wav
+D8_990 data_thchs30/test/D8_990.wav
+D8_991 data_thchs30/test/D8_991.wav
+D8_992 data_thchs30/test/D8_992.wav
+D8_993 data_thchs30/test/D8_993.wav
+D8_994 data_thchs30/test/D8_994.wav
+D8_995 data_thchs30/test/D8_995.wav
+D8_996 data_thchs30/test/D8_996.wav
+D8_997 data_thchs30/test/D8_997.wav
+D8_998 data_thchs30/test/D8_998.wav
+D8_999 data_thchs30/test/D8_999.wav

dict.txt

@@ -0,0 +1,1427 @@
+a1	阿啊呵腌吖锕
+a2	啊呵嗄
+a3	啊
+a4	啊呵
+a5	阿啊呵
+ai1	哀挨埃唉哎捱锿诶
+ai2	挨癌皑呆捱矮
+ai3	矮蔼哎霭嗳
+ai4	爱碍艾唉哎隘暧嗳瑷嗌嫒砹
+an1	安谙鞍氨庵桉鹌
+an3	俺铵揞埯
+an4	案按暗岸黯胺犴
+ang1	肮
+ang2	昂
+ang4	盎
+ao1	凹熬
+ao2	熬敖嚣嗷鏖鳌翱獒聱螯廒遨
+ao3	袄拗媪
+ao4	奥澳傲懊坳拗骜岙鏊
+ba1	八巴吧叭芭扒疤笆粑岜
+ba2	拔跋八茇菝魃
+ba3	把靶钯
+ba4	爸罢霸坝把耙灞鲅壩
+ba5	吧罢叭
+bai1	掰
+bai2	白百
+bai3	百摆伯柏佰捭
+bai4	败拜呗稗
+ban1	般班搬斑颁扳瘢癍
+ban3	版板阪坂钣舨
+ban4	办半伴扮瓣拌绊
+bang1	帮邦浜梆
+bang3	膀榜绑
+bang4	棒膀傍磅谤镑蚌蒡
+bao1	包胞炮剥褒苞孢煲龅
+bao2	薄雹保
+bao3	保宝饱堡葆褓鸨
+bao4	报暴抱爆鲍曝刨瀑豹趵
+bei1	背悲杯碑卑陂埤萆鹎
+bei3	北
+bei4	被备背辈倍贝蓓惫悖狈焙邶钡孛碚褙鐾鞴
+bei5	臂呗
+ben1	奔贲锛
+ben3	本苯畚
+ben4	奔笨夯坌
+beng1	崩绷嘣
+beng2	甭
+beng3	绷
+beng4	绷蹦迸甏泵蚌
+bi1	逼
+bi2	鼻荸匕比
+bi3	比笔彼鄙匕俾妣吡秕舭
+bi4	必毕币秘避闭壁臂弊辟碧拂毙蔽庇璧敝泌陛弼篦婢愎痹铋裨髀哔
+bian1	编边鞭砭
+bian3	贬扁匾碥褊
+bian4	便变遍辩辨辫卞苄汴忭弁缏
+bian5	边
+biao1	标彪镖膘骠镳杓飚飑飙瘭髟
+biao3	表裱婊
+biao4	鳔
+bie1	憋瘪鳖
+bie2	别蹩
+bie3	瘪
+bie4	别
+bin1	宾滨彬斌缤濒槟傧玢豳镔
+bin4	鬓殡摈膑髌
+bing1	兵冰并槟
+bing3	饼屏丙柄秉炳禀邴
+bing4	并病摒
+bo1	波播拨剥玻饽菠钵趵鲅
+bo2	博伯勃薄泊柏驳魄脖搏膊舶礴帛铂箔渤钹孛亳鹁踣饽
+bo3	簸跛
+bo4	薄柏簸擘檗
+bo5	卜啵膊
+bu1	逋晡钸
+bu2	不醭补
+bu3	补捕堡卜哺卟
+bu4	不部布步怖簿埔埠瓿钚
+ca1	擦拆嚓
+ca3	礤
+cai1	猜
+cai2	才财材裁
+cai3	采彩踩睬
+cai4	采菜蔡
+can1	参餐骖
+can2	残惭蚕
+can3	惨黪
+can4	灿掺璨孱粲
+cang1	苍仓沧舱伧
+cang2	藏
+cao1	操糙
+cao2	曹槽嘈漕螬艚草
+cao3	草
+ce4	策测侧厕册恻
+cen1	参
+cen2	岑涔
+ceng1	噌
+ceng2	曾层
+ceng4	蹭
+cha1	差插叉碴喳嚓杈馇锸
+cha2	查察茶叉茬碴楂猹搽槎檫
+cha3	叉衩镲
+cha4	差刹叉诧岔衩杈汊姹
+chai1	差拆钗
+chai2	柴豺侪
+chai4	虿瘥
+chan1	搀掺觇
+chan2	缠禅蝉馋潺蟾婵谗廛孱镡澶躔产单
+chan3	产铲阐谄冁蒇骣
+chan4	颤忏羼
+chang1	昌娼猖伥阊菖鲳
+chang2	长场常尝肠偿倘裳嫦徜苌厂
+chang3	场厂敞氅昶惝
+chang4	唱畅倡怅鬯
+chao1	超抄吵钞绰剿焯怊
+chao2	朝潮嘲巢晁
+chao3	炒吵
+chao4	耖
+che1	车砗
+che3	扯
+che4	彻撤澈掣坼辙
+chen1	郴琛嗔抻
+chen2	陈沉晨沈尘臣辰橙忱谌宸陳
+chen3	碜
+chen4	称趁衬谶榇龀
+chen5	伧
+cheng1	称撑秤瞠噌铛柽蛏
+cheng2	成城程承诚盛乘呈惩澄橙丞埕枨塍铖裎酲
+cheng3	逞骋裎
+cheng4	称秤
+chi1	吃痴哧嗤蚩笞鸱媸螭眵魑
+chi2	持迟池驰匙弛踟墀茌篪坻
+chi3	尺齿耻侈褫豉
+chi4	赤斥翅啻炽敕叱饬傺彳瘛眙
+chong1	冲充涌憧忡艟舂茺衝
+chong2	重崇虫
+chong3	宠
+chong4	冲铳
+chou1	抽瘳
+chou2	愁仇筹酬绸踌惆畴稠帱俦雠
+chou3	丑瞅
+chou4	臭
+chu1	出初樗
+chu2	除厨躇橱雏锄蜍刍滁蹰处
+chu3	处楚储础杵褚楮
+chu4	处触畜矗怵搐绌黜亍憷
+chuai1	揣搋
+chuai3	揣
+chuai4	揣啜踹嘬膪
+chuan1	穿川巛氚
+chuan2	传船遄椽舡
+chuan3	喘舛
+chuan4	串钏
+chuang1	创窗疮
+chuang2	床幢
+chuang3	闯
+chuang4	创怆
+chui1	吹炊
+chui2	垂锤捶陲椎槌棰
+chun1	春椿蝽
+chun2	纯唇醇淳鹑莼
+chun3	蠢
+chuo1	戳踔
+chuo4	绰啜辍龊
+ci1	差刺疵呲
+ci2	词辞慈磁瓷兹茨雌祠茈鹚糍
+ci3	此
+ci4	次刺赐伺蚝
+cong1	匆聪葱囱苁骢璁枞
+cong2	从丛琮淙
+cou4	凑辏腠
+cu1	粗
+cu2	徂殂
+cu4	促簇醋卒猝蹴蹙蔟酢
+cuan1	蹿撺汆镩
+cuan2	攒
+cuan4	窜篡爨
+cui1	衰催摧崔隹榱
+cui3	璀
+cui4	脆粹萃翠瘁悴淬毳啐
+cun1	村皴
+cun2	存蹲
+cun3	忖
+cun4	寸
+cuo1	搓撮磋蹉
+cuo2	嵯矬痤瘥鹾
+cuo3	脞
+cuo4	错措挫厝锉
+da1	答搭嗒耷褡哒
+da2	打达答瘩沓鞑怛笪靼妲
+da3	打
+da4	大
+da5	塔疸瘩
+dai1	待呆呔
+dai3	逮歹傣
+dai4	大代带待戴袋贷逮殆黛怠玳岱迨骀绐埭甙帶
+dan1	单担丹耽眈殚箪儋瘅聃郸
+dan3	担胆掸赕疸
+dan4	但担石弹淡旦蛋诞惮啖澹氮萏瘅
+dang1	当裆铛當
+dang3	党挡谠档
+dang4	当荡档挡宕菪凼砀
+dao1	刀叨忉氘
+dao2	叨导
+dao3	导倒岛蹈捣祷
+dao4	到道倒悼盗稻
+de2	得德锝嘚
+de4	嘚
+de5	的地得
+dei3	得
+deng1	登灯蹬噔簦
+deng3	等戥
+deng4	邓凳瞪澄蹬磴镫嶝
+di1	提低滴堤嘀氐镝羝
+di2	敌笛的迪涤嘀狄嫡翟荻籴觌镝
+di3	底抵诋邸砥坻柢氐骶
+di4	地第帝弟递蒂缔谛睇棣娣碲
+di5	弟
+dia3	嗲
+dian1	颠滇掂癫巅
+dian3	点典碘踮丶點
+dian4	电店甸淀垫殿奠惦佃玷簟坫靛钿癜阽
+diao1	雕刁凋叼貂碉鲷
+diao3	鸟屌
+diao4	调掉吊钓铫铞調
+die1	爹跌
+die2	叠迭碟谍蝶喋佚牒耋蹀堞瓞揲垤鲽
+ding1	丁盯钉叮町酊疔仃耵玎
+ding3	顶鼎酊
+ding4	定订钉铤腚锭碇啶
+diu1	丢铥丟
+dong1	东冬咚岽氡鸫
+dong3	懂董
+dong4	动洞冻栋恫侗垌峒胨胴硐動
+dou1	都兜蔸篼
+dou3	斗抖陡蚪
+dou4	读斗豆逗窦痘
+du1	都督嘟
+du2	读独顿毒渎牍犊黩髑椟
+du3	肚睹堵赌笃
+du4	度渡肚杜妒镀芏蠹
+duan1	端
+duan3	短
+duan4	断段锻缎煅椴簖
+dui1	堆
+dui4	对队兑敦碓憝怼镦對
+dun1	吨敦蹲墩礅镦
+dun3	盹趸
+dun4	顿盾钝炖遁沌囤砘
+duo1	多咄哆掇裰
+duo2	度夺踱铎掇
+duo3	朵躲垛哚缍
+duo4	舵堕跺剁惰垛驮柁
+e1	阿婀屙
+e2	额俄哦鹅娥峨蛾讹莪锇
+e3	恶
+e4	恶饿扼愕遏噩呃厄鄂轭颚鳄谔锷萼腭垩鹗苊阏惡
+e5	呃
+ei2	诶
+ei3	诶
+ei4	诶
+en1	恩蒽
+en4	摁嗯
+er2	而儿鸸鲕耳
+er3	尔耳迩饵洱珥铒
+er4	二佴
+fa1	发
+fa2	罚乏伐阀筏垡
+fa3	法砝
+fa4	发珐
+fan1	翻番帆藩幡蕃
+fan2	凡烦繁樊蕃燔矾蘩钒蹯
+fan3	反返
+fan4	饭犯范贩泛梵畈飯
+fang1	方芳妨坊邡枋钫
+fang2	房防妨坊肪鲂
+fang3	访仿纺彷舫紡
+fang4	放
+fang5	坊
+fei1	非飞啡菲扉霏妃绯蜚鲱
+fei2	肥腓淝
+fei3	菲匪诽斐蜚翡悱篚榧
+fei4	费废沸肺吠痱狒镄芾砩
+fen1	分纷氛芬吩酚玢
+fen2	坟焚汾棼鼢
+fen3	粉
+fen4	分份奋愤粪忿偾瀵鲼
+feng1	风封丰峰疯锋蜂枫烽酆葑沣砜鋒
+feng2	逢缝冯
+feng3	讽唪
+feng4	奉缝凤俸葑鳳
+fo2	佛
+fou3	否缶
+fu1	夫肤敷孵呋稃麸趺跗
+fu2	夫服福佛幅伏符浮扶弗拂袱俘芙孚匐辐涪氟桴蜉苻茯莩菔幞怫艴郛绂绋凫祓砩黻罘稃蚨芾蝠
+fu3	府父腐抚辅甫俯斧脯釜腑拊滏黼符莆
+fu4	服复父负副富付妇附赴腹覆赋傅缚咐阜讣驸赙馥蝮鲋鳆負
+fu5	咐服袱夫腐复
+ga1	咖嘎胳伽旮
+ga2	嘎噶轧尜钆
+ga3	嘎尕
+ga4	尬
+gai1	该赅垓陔
+gai3	改
+gai4	概盖丐钙芥溉戤
+gan1	干甘肝杆尴乾竿坩苷柑泔矸疳酐
+gan2	赶敢
+gan3	感敢赶杆橄秆擀澉
+gan4	干赣淦绀旰
+gang1	刚钢纲缸扛杠冈肛罡剛
+gang3	港岗
+gang4	钢杠戆筻
+gao1	高糕膏皋羔睾篙槔
+gao3	稿搞藁槁缟镐杲
+gao4	告膏诰郜锆
+ge1	歌格哥戈割胳搁疙咯鸽屹仡圪纥袼
+ge2	革格隔葛阁胳搁蛤嗝骼颌搿膈镉塥鬲噶
+ge3	个各合盖葛哿舸
+ge4	个各铬硌虼個
+gei3	给給
+gen1	根跟
+gen2	哏
+gen3	艮
+gen4	亘艮茛
+geng1	更耕庚羹赓
+geng3	耿颈梗哽鲠埂绠
+geng4	更
+gong1	工公共红供功攻宫恭躬龚弓肱蚣觥
+gong3	巩拱汞珙
+gong4	共供贡
+gou1	句沟勾钩篝佝枸缑鞲
+gou3	狗苟岣枸笱
+gou4	够购构勾觏垢诟媾遘彀
+gu1	姑骨孤估辜咕呱箍沽菇轱鸪毂菰蛄酤觚
+gu2	骨古鼓谷
+gu3	古股鼓骨谷贾汩蛊毂鹄牯臌诂瞽罟钴嘏蛄鹘
+gu4	故顾固估雇锢梏牿崮痼鲴
+gua1	括瓜刮呱栝胍鸹
+gua3	寡呱剐
+gua4	挂褂卦诖
+guai1	乖掴
+guai3	拐
+guai4	怪
+guan1	关观官冠棺矜莞倌纶鳏
+guan3	管馆莞
+guan4	观惯冠贯罐灌掼盥涫鹳
+guang1	光咣胱桄
+guang3	广犷
+guang4	逛桄
+gui1	规归瑰龟硅闺皈傀圭妫鲑
+gui3	鬼轨诡癸匦庋宄晷簋
+gui4	贵桂跪柜刽炔刿桧炅鳜匮
+gun3	滚鲧衮绲磙辊
+gun4	棍
+guo1	过锅郭涡聒蝈崞埚呙
+guo2	国帼掴馘虢
+guo3	果裹猓椁蜾
+guo4	过過
+ha1	哈铪
+ha2	虾蛤
+ha3	哈
+ha4	哈
+hai1	嘿咳嗨
+hai2	还孩骸海還
+hai3	海胲醢
+hai4	害亥骇氦
+han1	酣憨顸鼾蚶
+han2	含寒汗韩涵函晗焓邯邗
+han3	喊罕阚
+han4	汉汗憾翰撼旱捍悍瀚焊颔菡撖
+hang1	夯
+hang2	行航吭杭绗珩颃
+hang4	行巷沆
+hao1	蒿薅嚆
+hao2	号毫豪嚎壕貉嗥濠蚝好
+hao3	好郝
+hao4	好号浩耗皓昊灏镐颢
+he1	喝呵诃嗬
+he2	和何合河核盒禾荷阂涸阖貉曷颌劾菏盍纥蚵翮
+he4	和何喝赫吓贺荷鹤壑褐
+hei1	黑嘿嗨
+hen2	痕很
+hen3	很狠
+hen4	恨
+heng1	哼亨
+heng2	横衡恒蘅珩桁
+heng4	横
+heng5	哼
+hong1	轰哄烘薨訇
+hong2	红洪鸿宏虹弘泓闳蕻黉荭
+hong3	哄
+hong4	哄讧蕻
+hou2	侯喉猴瘊篌糇骺
+hou3	吼
+hou4	后候後厚侯逅堠鲎
+hu1	乎呼戏忽糊惚唿滹轷烀
+hu2	和胡湖糊核壶狐葫弧蝴囫瑚斛鹄醐猢槲鹕觳煳鹘
+hu3	虎浒唬琥
+hu4	护户互糊虎沪祜扈戽笏岵怙瓠鹱冱
+hua1	华化花哗砉
+hua2	华划滑哗豁猾骅铧
+hua4	话华化划画桦
+huai2	怀徊淮槐踝
+huai4	坏
+huai5	划
+huan1	欢獾
+huan2	还环寰鬟桓圜洹郇缳锾萑
+huan3	缓
+huan4	换患幻唤宦焕痪涣浣奂擐豢漶逭鲩
+huang1	荒慌肓
+huang2	黄皇煌惶徨璜簧凰潢蝗蟥遑隍磺癀湟篁鳇黃
+huang3	晃恍谎幌
+huang4	晃
+hui1	挥辉灰恢徽堕诙晖麾珲咴虺隳
+hui2	回徊蛔茴洄
+hui3	毁悔虺
+hui4	会汇惠慧溃绘讳贿晦秽诲彗烩荟卉喙恚浍哕缋桧蕙蟪
+hun1	婚昏荤阍
+hun2	混魂浑馄珲
+hun4	混诨溷
+huo1	豁劐攉锪耠
+huo2	和活火
+huo3	火伙夥钬
+huo4	和或获货祸惑霍豁藿嚯镬蠖
+ji1	其几期机基击奇激积鸡迹绩饥缉圾姬矶肌讥叽稽畸跻羁嵇唧畿齑箕屐剞玑赍犄墼芨丌咭笄乩績
+ji2	及即辑级极��急籍吉疾嫉藉脊棘汲岌笈瘠诘亟楫蒺殛佶戢嵴蕺给圾
+ji3	几给己革济纪挤脊戟虮掎麂
+ji4	记系计济寄际技纪继既齐季寂祭忌剂冀妓骥蓟悸伎暨霁稷偈鲫髻觊荠跽哜鲚洎芰迹绩
+jia1	家加佳夹嘉茄挟枷珈迦伽浃痂笳葭镓袈跏
+jia2	夹颊戛荚郏恝铗袷蛱浃
+jia3	假甲贾胛岬钾嘏瘕
+jia4	价假架驾嫁稼
+jia5	家
+jian1	间坚监渐兼艰肩浅尖奸溅煎歼缄笺菅蒹搛湔缣戋犍鹣鲣鞯
+jian3	简减检剪捡拣俭碱茧柬蹇謇硷睑锏枧戬谫囝裥笕翦趼
+jian4	见间件建监渐健剑键荐鉴践舰箭贱溅槛谏僭涧饯毽锏楗腱牮踺
+jiang1	将江疆姜浆僵缰茳礓豇
+jiang3	讲奖蒋桨耩蔣
+jiang4	将强降酱浆虹匠犟绛洚糨
+jiao1	教交焦骄郊胶椒娇浇姣跤蕉礁鲛僬鹪蛟艽茭
+jiao2	嚼矫脚
+jiao3	角脚搅缴绞饺矫佼狡剿侥皎挢徼湫敫铰
+jiao4	教觉校叫较轿嚼窖酵噍峤徼醮
+jie1	接结节街阶皆揭楷嗟秸疖喈
+jie2	结节杰捷截洁劫竭睫桔拮孑诘桀碣偈颉讦婕羯鲒解結
+jie3	解姐
+jie4	界解价介借戒届藉诫芥疥蚧骱
+jie5	姐
+jin1	今金禁津斤筋巾襟矜衿
+jin3	尽仅紧谨锦瑾馑卺廑堇槿
+jin4	进近尽仅禁劲晋浸靳缙烬噤觐荩赆妗
+jing1	经京精惊睛晶荆兢鲸泾旌茎腈菁粳
+jing3	警景井颈憬阱儆刭肼
+jing4	经境竟静敬镜劲竞净径靖痉迳胫弪婧獍靓
+jiong1	扃
+jiong3	窘炯迥炅囧
+jiu1	究纠揪鸠赳啾阄鬏
+jiu3	九酒久韭灸玖
+jiu4	就旧救疚舅咎臼鹫僦厩桕柩
+ju1	车据且居俱拘驹鞠锯趄掬疽裾苴椐锔狙琚雎鞫蜛沮
+ju2	局菊桔橘锔举
+ju3	举柜矩咀沮踽龃榉莒枸
+ju4	据句具剧巨聚拒距俱惧沮瞿锯炬飓踞遽倨钜犋屦榘窭讵醵苣
+juan1	捐圈娟鹃涓镌蠲
+juan3	卷锩
+juan4	圈卷倦眷隽绢狷桊鄄劵
+jue1	嗟撅噘
+jue2	觉绝决角脚嚼掘诀崛爵抉倔獗厥蹶攫谲矍孓橛噱珏桷劂爝镢蕨觖
+jue3	蹶
+jue4	倔
+jun1	军均君钧筠龟菌皲麇
+jun4	俊峻隽菌郡骏竣捃浚
+ka1	咖喀咔
+ka3	卡咯咔佧胩
+kai1	开揩锎
+kai3	慨凯铠楷恺蒈剀垲锴
+kai4	忾
+kan1	看刊堪勘龛戡
+kan3	侃砍坎槛莰
+kan4	看嵌瞰阚磡
+kang1	康慷糠闶
+kang2	扛
+kang4	抗炕亢伉闶钪
+kao1	尻
+kao3	考烤拷栲
+kao4	靠铐犒
+ke1	科颗柯呵棵苛磕坷嗑瞌轲稞疴蝌钶窠颏珂髁
+ke2	咳壳颏可
+ke3	可渴坷轲岢
+ke4	克客刻课恪嗑溘骒缂氪锞蚵科可
+ken3	肯恳啃垦龈
+ken4	裉
+keng1	坑吭铿
+kong1	空倥崆箜
+kong3	恐孔倥
+kong4	空控
+kou1	抠芤眍
+kou3	口
+kou4	扣寇叩蔻筘
+ku1	哭枯窟骷刳堀
+ku3	苦
+ku4	库裤酷喾绔
+kua1	夸
+kua3	垮侉
+kua4	跨挎胯
+kuai3	蒯
+kuai4	快块筷脍会哙侩狯浍郐
+kuan1	宽髋
+kuan3	款
+kuang1	框筐匡哐诓
+kuang2	狂诳
+kuang3	夼
+kuang4	况矿框旷眶邝圹纩贶況
+kui1	亏窥盔岿悝
+kui2	魁睽逵葵奎馗夔喹隗暌揆蝰
+kui3	傀跬
+kui4	愧溃馈匮喟聩篑蒉愦
+kun1	昆坤鲲锟醌琨髡
+kun3	捆悃阃
+kun4	困
+kuo4	括适阔扩廓栝蛞
+la1	拉啦喇垃邋
+la2	拉喇旯砬
+la3	拉喇
+la4	落拉辣腊蜡剌瘌
+la5	啦
+lai2	来莱徕涞崃铼來
+lai4	赖睐癞籁赉濑
+lan2	兰蓝栏拦篮澜婪岚斓阑褴镧谰
+lan3	懒览揽榄缆漤罱懶
+lan4	烂滥
+lang1	啷
+lang2	狼郎廊琅螂榔锒稂阆
+lang3	朗
+lang4	浪郎莨蒗阆
+lao1	捞
+lao2	劳牢唠崂铹痨醪老
+lao3	老姥佬潦栳铑
+lao4	落络唠烙酪涝耢
+lao5	姥
+le1	肋
+le4	乐勒仂叻泐鳓
+le5	了
+lei1	勒擂
+lei2	累雷擂羸镭嫘缧檑蕾
+lei3	累蕾垒磊儡诔耒
+lei4	类泪累擂肋酹
+lei5	嘞
+leng1	棱
+leng2	楞棱塄冷
+leng3	冷
+leng4	愣
+li1	哩
+li2	离丽黎璃漓狸梨篱犁厘罹藜骊蜊黧缡喱鹂嫠蠡鲡蓠理礼離嚟
+li3	里理李礼哩鲤俚逦娌悝澧锂蠡醴鳢裡
+li4	力利立历例丽励厉莉笠粒俐栗隶吏沥雳莅戾俪砺痢郦詈荔枥呖唳猁溧砾栎轹傈坜苈疠疬蛎鬲篥粝跞
+li5	璃哩莉里
+lia3	俩
+lian2	联连怜莲廉帘涟镰裢濂臁奁蠊鲢
+lian3	脸敛琏蔹裣
+lian4	练恋炼链殓楝潋
+liang2	量良梁凉粮粱踉莨椋墚两
+liang3	两俩魉
+liang4	量亮辆凉谅晾踉靓
+liao1	撩撂
+liao2	聊疗辽僚寥撩撂缭寮燎嘹獠鹩了
+liao3	了潦燎蓼钌
+liao4	了料廖镣撩撂尥钌
+lie1	咧
+lie3	裂咧
+lie4	列烈裂劣猎趔冽洌捩埒躐鬣
+lie5	咧
+lin2	林临邻琳淋霖麟鳞磷嶙辚粼遴啉瞵
+lin3	凛懔檩廪
+lin4	淋吝躏赁蔺膦
+lin1	拎
+ling2	令灵零龄凌玲铃陵伶聆囹棱菱苓翎棂瓴绫酃泠羚蛉柃鲮领岭
+ling3	领令岭
+ling4	令另呤
+liu1	溜熘
+liu2	留流刘瘤榴浏硫琉遛馏镏旒骝鎏柳
+liu3	柳绺锍
+liu4	六溜碌遛馏镏鹨
+lo5	咯
+long1	隆
+long2	龙隆笼胧咙聋珑窿茏栊泷砻癃
+long3	笼拢垄陇垅
+long4	弄
+lou1	搂
+lou2	楼喽偻娄髅蝼蒌耧
+lou3	搂篓嵝
+lou4	露陋漏镂瘘
+lou5	喽
+lu1	噜撸
+lu2	卢炉庐芦颅泸轳鲈垆胪鸬舻栌
+lu3	鲁芦卤虏掳橹镥
+lu4	六路陆录露绿鹿碌禄辘麓赂漉戮簏鹭潞璐辂渌蓼逯
+lu5	轳氇噜
+lv3	旅履屡侣缕吕捋铝偻褛膂稆
+lv4	律绿率虑滤氯緑
+lv5	驴榈闾率
+luan2	峦挛孪栾銮滦鸾娈脔
+luan3	卵
+luan4	乱
+lve3	掠
+lve4	略掠锊
+lun1	抡
+lun2	论轮伦沦仑抡囵纶
+lun4	论
+luo1	落罗捋啰
+luo2	罗逻萝螺锣箩骡猡椤脶镙
+luo3	裸倮蠃瘰
+luo4	落络洛骆咯摞烙珞泺漯荦硌雒
+luo5	罗
+m2	呒
+ma1	妈麻摩抹蚂嬷
+ma2	吗麻蟆马玛
+ma3	马吗码玛蚂犸
+ma4	骂蚂唛杩犸
+ma5	么吗嘛妈嗎
+mai2	埋霾
+mai3	买荬買
+mai4	卖麦迈脉劢
+man1	颟
+man2	蛮馒瞒埋蔓谩鳗鞔满
+man3	满螨
+man4	慢漫曼蔓谩墁幔缦熳镘
+mang2	忙茫盲芒氓邙硭
+mang3	莽蟒漭
+mao1	猫貓
+mao2	毛猫矛茅髦锚牦旄蝥蟊茆
+mao3	卯铆峁泖昴
+mao4	冒贸帽貌茂耄瑁懋袤瞀貿
+me5	么麽麼
+mei2	没眉梅媒枚煤霉玫糜酶莓嵋湄楣猸镅鹛美每沒
+mei3	美每镁浼
+mei4	妹魅昧谜媚寐袂
+men1	闷
+men2	门扪钔
+men4	闷懑焖
+men5	们們
+meng1	蒙
+meng2	蒙盟朦氓萌檬瞢甍礞虻艨濛
+meng3	蒙猛勐懵蠓蜢锰艋
+meng4	梦孟
+mi1	眯咪
+mi2	迷弥谜靡糜醚麋猕祢縻蘼米
+mi3	米眯靡弭敉脒芈
+mi4	密秘觅蜜谧泌汨宓幂嘧
+mian2	棉眠绵
+mian3	免缅勉腼冕娩渑湎沔眄黾
+mian4	面
+miao1	喵
+miao2	描苗瞄鹋
+miao3	秒渺藐缈淼杪邈眇
+miao4	妙庙缪
+mie1	乜咩
+mie4	灭蔑篾蠛
+min2	民珉岷缗玟苠
+min3	敏悯闽泯皿抿闵愍黾鳘
+ming2	名明鸣盟铭冥茗溟瞑暝螟
+ming3	酩
+ming4	命
+miu4	谬缪
+mo1	摸
+mo2	无模麽磨摸摩魔膜蘑馍摹
+mo3	抹
+mo4	莫陌磨末默沫漠寞没墨茉
+mo5	寞沫
+mu3	拇牡亩姆母姥
+mu4	墓暮幕募慕木目睦牧穆沐
+na2	拿哪
+na3	哪
+na4	呐钠那娜纳衲
+na5	呐
+nai3	氖乃奶妳
+nai4	耐奈
+nan2	南男难楠喃
+nan3	腩蝻赧
+nan4	难
+nang2	囊
+nao1	孬
+nao2	挠
+nao3	脑恼
+nao4	闹淖
+ne2	哪
+ne5	呢
+nei3	馁
+nei4	内
+nen4	嫩
+neng2	能
+ni1	妮
+ni2	霓倪泥尼
+ni3	拟你
+ni4	匿腻逆溺昵
+nian1	蔫拈
+nian2	年粘黏拈鲶
+nian3	碾撵捻
+nian4	念
+niang2	娘
+niang4	酿
+niao3	鸟袅嬲
+niao4	尿
+nie1	捏㖏
+nie4	聂孽啮镊镍涅
+nin2	您
+ning2	柠狞凝宁拧咛
+ning3	拧
+ning4	拧泞宁
+niu2	牛
+niu3	扭钮纽
+niu4	拗
+nong1	浓脓
+nong2	浓农侬
+nong4	弄
+nu2	奴孥
+nu3	努弩
+nu4	怒
+nv3	女钕
+nuan3	暖
+nue4	虐疟
+nuo2	挪傩娜
+nuo4	懦糯诺
+o2	哦
+o5	哦
+ou1	欧鸥殴瓯呕歐
+ou3	藕呕偶
+ou4	沤
+pa1	啪趴爬葩
+pa2	爬琶杷
+pa4	帕怕
+pa5	琶杷
+pai1	拍
+pai2	排牌徘
+pai4	湃派
+pan1	攀潘
+pan2	盘胖磐
+pan4	盼畔判叛
+pang1	乓
+pang2	庞旁螃彷
+pang3	耪
+pang4	胖
+pao1	抛
+pao2	咆袍刨庖
+pao3	跑
+pao4	炮泡
+pei1	呸胚
+pei2	培裴赔陪
+pei4	配佩沛
+pen1	喷
+pen2	盆
+peng1	砰抨烹
+peng2	澎彭蓬棚硼篷膨朋鹏芃
+peng3	捧
+peng4	碰
+peng5	篷蓬
+pi1	坯砒霹批披劈丕
+pi2	皮琵毗啤脾疲枇铍貔
+pi3	匹痞癖
+pi4	僻屁譬辟媲
+pian1	篇偏片翩
+pian2	骈
+pian4	片骗
+piao1	飘漂
+piao2	瓢朴嫖漂
+piao3	瞟
+piao4	漂票嫖
+pie1	撇瞥
+pie3	撇瞥
+pin1	拼拚
+pin2	频贫品嫔
+pin3	品
+pin4	聘
+ping1	乒娉
+ping2	坪苹萍平凭瓶评屏
+po1	坡泼颇
+po2	婆鄱
+po3	颇叵
+po4	破魄迫粕朴珀
+pou1	剖
+pu1	扑铺仆噗
+pu2	仆莆葡菩蒲谱濮璞
+pu3	普谱埔朴圃浦
+pu4	铺曝瀑
+qi1	期七欺栖漆妻凄柒沏戚桼
+qi2	其旗琪骑骐棋歧岐琦齐奇祁祺麒起祈淇芪锜脐綦鳍圻
+qi3	起启企岂祈乞绮杞
+qi4	弃气器汽戚泣砌契讫
+qia1	掐葜袷恰
+qia3	卡
+qia4	恰洽髂
+qian1	千签牵迁谦铅骞悭芊愆阡仟岍扦佥搴褰钎
+qian2	前钱潜乾虔钳掮黔荨钤犍箝
+qian3	浅遣谴缱肷
+qian4	欠歉纤嵌倩堑茜芡慊椠
+qiang1	将枪抢腔呛锵跄羌戕戗镪蜣锖
+qiang2	强墙蔷樯嫱
+qiang3	强抢襁镪羟
+qiang4	呛跄炝戗
+qiao1	悄敲雀锹跷橇缲硗劁
+qiao2	桥乔侨瞧翘蕉憔樵峤谯荞鞒
+qiao3	悄巧雀愀
+qiao4	翘俏窍壳峭撬鞘诮谯
+qie1	切
+qie3	且
+qie4	切窃怯趄妾砌惬锲挈郄箧慊
+qin1	亲钦侵衾
+qin2	琴秦勤芹擒矜覃禽噙廑溱檎嗪芩螓
+qin3	寝
+qin4	沁揿吣
+qing1	青清轻倾卿氢蜻圊鲭
+qing2	情晴擎氰檠黥
+qing3	请顷謦苘
+qing4	亲庆罄磬箐綮
+qiong2	穷琼穹茕邛蛩筇跫銎
+qiu1	秋邱丘龟蚯鳅楸湫
+qiu2	求球仇囚酋裘虬俅遒赇泅逑犰蝤巯鼽
+qiu3	糗
+qu1	区曲屈趋驱躯觑岖蛐祛蛆麴诎黢
+qu2	渠瞿衢癯劬璩氍朐磲鸲蕖蠼蘧取佢
+qu3	取曲娶龋苣
+qu4	去趣觑阒
+qu5	戌去
+quan1	圈悛
+quan2	全权泉拳诠颧蜷荃铨痊醛辁筌鬈權
+quan3	犬绻畎
+quan4	劝券
+que1	缺阙炔
+que2	瘸
+que4	却确雀榷鹊阕阙悫
+qun1	逡
+qun2	群裙麇
+ran2	然燃髯蚺
+ran3	染冉苒
+rang1	嚷
+rang2	瓤禳穰
+rang3	嚷攘壤
+rang4	让
+rao2	饶娆桡荛
+rao3	扰绕娆
+rao4	绕
+re3	若惹喏
+re4	热
+ren2	人任仁壬
+ren3	忍稔荏
+ren4	任认韧刃纫饪仞葚妊轫衽恁
+ren5	人
+reng1	扔
+reng2	仍
+ri4	日
+rong2	容荣融蓉溶绒熔榕戎嵘茸狨肜蝾
+rong3	冗
+rou2	柔揉蹂糅鞣
+rou4	肉
+ru2	如儒茹嚅濡孺蠕薷铷襦颥乳
+ru3	辱乳汝
+ru4	入褥缛洳溽蓐
+ruan3	软���朊
+rui2	蕤
+rui3	蕊
+rui4	瑞锐芮睿枘蚋
+run4	润闰
+ruo4	若弱偌箬
+sa1	撒仨挲
+sa3	洒撒
+sa4	萨卅飒脎
+sai1	思塞腮鳃噻
+sai4	赛塞
+san1	三毵
+san3	散伞馓糁霰
+san4	散
+sang1	丧桑
+sang3	嗓搡磉颡
+sang4	丧
+sao1	骚搔臊缲缫鳋
+sao3	扫嫂
+sao4	扫梢臊埽瘙
+se4	色塞涩瑟啬铯穑
+sen1	森
+seng1	僧
+sha1	杀沙刹纱杉莎煞砂挲鲨痧裟铩
+sha3	傻
+sha4	沙啥厦煞霎嗄歃唼
+shai1	筛酾
+shai3	色
+shai4	晒
+shan1	山衫删煽扇珊杉栅跚姗潸膻芟埏钐舢苫髟
+shan3	闪陕掺
+shan4	单善扇禅擅膳讪汕赡缮嬗掸骟剡苫鄯钐疝蟮鳝
+shang1	商伤汤殇觞熵墒
+shang3	上赏晌垧
+shang4	上尚绱
+shang5	裳上
+shao1	烧稍梢捎鞘蛸筲艄
+shao2	勺韶苕杓芍
+shao3	少
+shao4	少绍召稍哨邵捎潲劭
+she1	奢赊猞畲
+she2	折舌蛇佘
+she3	舍
+she4	社设舍涉射摄赦慑麝滠歙厍
+shei2	谁
+shen1	身深参申伸绅呻莘娠诜砷糁珅
+shen2	什神
+shen3	审沈婶谂哂渖矧
+shen4	甚慎渗肾蜃葚胂椹
+sheng1	生声胜升牲甥笙
+sheng2	绳渑
+sheng3	省眚
+sheng4	胜圣盛乘剩嵊晟
+shi1	师诗失施尸湿狮嘘虱蓍酾鲺
+shi2	十时实什识食石拾蚀埘莳炻鲥
+shi3	使始史驶屎矢豕
+shi4	是事世市士式视似示室势试释适氏饰逝誓嗜侍峙仕恃柿轼拭噬弑谥莳贳铈螫舐筮识
+shi5	殖匙识
+shou1	收
+shou2	熟手首
+shou3	手首守艏
+shou4	受授售瘦寿兽狩绶
+shu1	书输殊舒叔疏抒淑梳枢蔬倏菽摅姝纾毹殳疋
+shu2	熟孰赎塾秫
+shu3	数属署鼠薯暑蜀黍曙
+shu4	数术树述束竖恕墅漱俞戍庶澍沭腧
+shua1	刷唰
+shua3	耍
+shua4	刷
+shuai1	衰摔
+shuai3	甩
+shuai4	率帅蟀
+shuan1	栓拴闩
+shuan4	涮
+shuang1	双霜孀泷
+shuang3	爽
+shui2	谁水
+shui3	水
+shui4	税睡说
+shun3	吮
+shun4	顺舜瞬
+shuo1	说
+shuo4	数朔硕烁铄妁蒴槊搠
+si1	思斯司私丝撕厮嘶鸶咝澌缌锶蛳
+si3	死
+si4	四似食寺肆伺饲嗣巳祀驷泗俟汜兕姒耜笥
+si5	司思
+song1	松忪淞崧嵩凇菘
+song3	耸悚怂竦
+song4	送宋诵颂讼
+sou1	搜艘馊嗖溲飕锼螋
+sou3	擞叟薮嗾瞍
+sou4	嗽擞
+su1	苏稣酥
+su2	俗
+su4	诉速素肃宿缩塑溯粟簌夙嗉谡僳愫涑蔌觫
+suan1	酸狻
+suan4	算蒜
+sui1	虽尿荽睢眭濉
+sui2	随遂隋绥
+sui3	髓
+sui4	岁碎遂祟隧邃穗燧谇
+sun1	孙荪狲飧
+sun3	损笋榫隼
+suo1	缩莎梭嗦唆挲娑睃桫嗍蓑羧
+suo2	所索
+suo3	所索锁琐唢
+suo5	嗦
+ta1	他她它踏塌遢溻铊趿
+ta3	塔鳎獭
+ta4	踏拓榻嗒蹋沓挞闼漯遢
+tai1	台胎苔
+tai2	台抬苔邰薹骀炱跆鲐
+tai3	呔
+tai4	太态泰汰酞肽钛
+tan1	摊贪滩瘫坍
+tan2	谈弹坛谭潭覃痰澹檀昙锬镡郯
+tan3	坦毯忐袒钽
+tan4	探叹炭碳
+tan5	弹
+tang1	汤趟铴镗耥羰
+tang2	堂唐糖膛塘棠搪溏螳瑭樘镗螗饧醣
+tang3	躺倘淌傥帑
+tang4	趟烫
+tao1	涛掏滔叨焘韬饕绦
+tao2	逃陶桃淘萄啕洮鼗
+tao3	讨
+tao4	套
+te4	特忑忒慝铽
+tei1	忒
+teng2	腾疼藤誊滕
+ti1	体踢梯剔锑
+ti2	提题啼蹄醍绨缇鹈荑
+ti3	体
+ti4	替涕剃惕屉嚏悌倜逖绨裼
+tian1	天添
+tian2	田填甜恬佃阗畋钿嗔
+tian3	腆舔忝殄
+tian4	掭
+tiao1	挑佻祧
+tiao2	条调迢鲦苕髫龆蜩笤
+tiao3	挑窕
+tiao4	跳眺粜
+tie1	贴帖萜
+tie3	铁帖
+tie4	帖餮
+ting1	听厅汀烃
+ting2	停庭亭婷廷霆蜓葶莛
+ting3	挺艇町铤梃
+ting4	梃
+tong1	通恫嗵
+tong2	同童彤铜桐瞳佟酮侗仝垌茼峒潼砼硐
+tong3	统筒桶捅侗
+tong4	同通痛恸
+tou1	偷
+tou2	头投骰
+tou3	钭
+tou4	透
+tu1	突秃凸
+tu2	图途徒屠涂荼菟酴凃
+tu3	土吐钍
+tu4	吐兔堍菟
+tuan1	湍
+tuan2	团抟
+tuan3	疃
+tuan4	彖
+tui1	推忒
+tui2	颓
+tui3	腿
+tui4	退褪蜕煺
+tun1	吞暾
+tun2	屯饨臀囤豚炖
+tun3	氽
+tun4	饨
+tuo1	托脱拖乇
+tuo2	陀舵驼砣驮沱跎坨鸵橐佗铊酡柁鼍
+tuo3	妥椭庹
+tuo4	魄拓唾柝箨
+wa1	挖哇凹娲蛙洼
+wa2	娃瓦
+wa3	瓦佤
+wa4	瓦袜腽
+wa5	哇娃
+wai1	歪
+wai3	崴
+wai4	外
+wan1	湾弯蜿剜豌灣
+wan2	完玩顽丸纨芄烷碗婉
+wan3	晚碗挽婉惋宛莞娩畹皖绾琬脘菀
+wan4	万腕蔓
+wang1	汪
+wang2	王亡芒往网
+wang3	往网枉惘罔辋魍
+wang4	望王往忘旺妄
+wei1	威微危巍萎偎薇逶煨崴葳隈委
+wei2	为维围唯违韦惟帷帏圩囗潍桅嵬闱沩涠位薇
+wei3	委伟唯尾玮伪炜纬萎娓苇猥痿韪洧隗诿艉鲔
+wei4	为位未味卫谓遗慰魏蔚畏胃喂尉渭猬軎
+wen1	温瘟
+wen2	文闻纹蚊雯璺阌稳炆
+wen3	稳吻紊刎
+wen4	问纹汶璺紊搵
+weng1	翁嗡
+weng3	蓊
+weng4	瓮蕹
+wo1	窝涡蜗喔倭挝莴
+wo2	哦我
+wo3	我
+wo4	握卧渥沃斡幄肟硪龌
+wu1	於恶屋污乌巫呜诬兀钨邬圬
+wu2	无亡吴吾捂毋梧唔芜浯蜈五舞無
+wu3	五武午舞伍侮捂妩忤鹉牾迕庑怃仵
+wu4	物务误恶悟乌雾勿坞戊兀晤鹜痦寤骛芴杌焐阢婺鋈
+xi1	西息希吸惜稀悉析夕牺腊昔熙兮溪嘻锡晰樨熄膝栖郗犀曦奚羲唏蹊淅皙汐嬉茜熹烯翕蟋歙浠僖穸蜥螅菥舾矽粞硒醯欷鼷
+xi2	席习袭媳檄隰觋
+xi3	喜洗禧徙玺屣���蓰铣
+xi4	系细戏隙饩阋禊舄
+xia1	瞎虾呷
+xia2	峡侠狭霞暇辖遐匣黠瑕狎硖柙
+xia4	下夏吓厦唬
+xian1	先鲜仙掀纤暹莶锨氙祆籼酰跹
+xian2	闲贤嫌咸弦娴衔涎舷鹇痫閑
+xian3	显险鲜洗跣猃藓铣燹蚬筅冼癣
+xian4	现见线限县献宪陷羡馅腺岘苋霰
+xiang1	相香乡箱厢湘镶襄骧葙芗缃
+xiang2	降详祥翔庠想
+xiang3	想响享飨饷鲞
+xiang4	相向象像项巷橡蟓
+xiao1	消销潇肖萧宵削嚣逍硝霄哮枭骁箫枵哓蛸绡魈銷
+xiao2	淆崤小
+xiao3	小晓筱
+xiao4	笑校效肖孝啸哮
+xie1	些歇楔蝎
+xie2	叶协鞋携斜胁谐邪挟偕撷勰颉缬写
+xie3	写血
+xie4	写解谢泄契械屑卸懈泻亵蟹邂榭瀣薤燮躞廨绁渫榍獬
+xin1	心新欣辛薪馨鑫芯昕忻歆锌
+xin2	寻镡
+xin4	信芯衅
+xing1	兴星腥惺猩
+xing2	行形型刑邢陉荥饧硎
+xing3	省醒擤
+xing4	性兴姓幸杏悻荇
+xiong1	兄胸凶匈汹芎
+xiong2	雄熊
+xiu1	修休羞咻馐庥鸺貅髹貅
+xiu3	宿朽
+xiu4	秀袖宿臭绣锈嗅岫溴
+xu1	需须虚吁嘘墟戌胥砉圩盱顼虛
+xu2	徐
+xu3	许浒栩诩糈醑
+xu4	续序绪蓄叙畜恤絮旭婿酗煦洫溆勖
+xu5	蓿
+xuan1	宣喧轩萱暄谖揎儇煊瑄
+xuan2	旋悬玄漩璇痃选玹
+xuan3	选癣
+xuan4	旋券炫渲绚眩铉泫碹楦镟
+xue1	削靴薛
+xue2	学穴噱踅泶
+xue3	雪鳕
+xue4	血谑
+xun1	熏勋荤醺薰埙曛窨獯
+xun2	寻询巡循旬驯荀峋洵恂郇浔鲟荨
+xun4	训迅讯逊熏殉巽徇汛蕈浚
+ya1	压雅呀押鸦哑鸭丫垭桠
+ya2	牙涯崖芽衙睚伢岈琊蚜
+ya3	雅痖疋哑
+ya4	亚压讶轧娅迓揠氩砑
+ya5	呀
+yan1	烟燕咽殷焉淹阉腌嫣胭湮阏鄢菸崦恹
+yan2	言严研延沿颜炎阎盐岩蜒檐妍筵芫闫阽眼
+yan3	眼演掩衍奄俨偃魇鼹兖郾琰罨厣剡
+yan4	验厌燕宴咽雁焰艳谚彦研沿焱晏唁砚堰赝餍滟酽谳
+yang1	央泱秧鸯殃鞅
+yang2	洋阳杨扬羊疡佯烊徉炀蛘养仰
+yang3	养仰痒氧
+yang4	样漾恙烊怏鞅
+yao1	要约邀腰夭妖吆幺么
+yao2	摇遥姚陶尧谣瑶窑肴侥铫珧轺爻徭繇鳐垚
+yao3	咬杳窈舀崾
+yao4	要药耀钥鹞曜疟
+ye1	耶噎椰掖
+ye2	爷耶邪揶铘也野
+ye3	也野冶
+ye4	业夜叶页液咽曳拽烨掖腋谒邺靥晔
+yi1	一医衣依椅伊漪咿揖噫猗铱欹黟
+yi2	移疑遗宜仪蛇姨夷怡颐彝咦贻迤痍胰沂饴圯荑诒眙嶷一以谊
+yi3	以已衣尾椅矣乙蚁倚迤蛾旖苡钇舣酏
+yi4	意义议易衣艺译异益亦亿忆谊抑翼役艾溢毅裔逸轶弈翌疫绎佚奕熠诣弋驿懿呓屹薏噫镒缢邑臆刈羿仡峄怿悒肄佾殪挹埸劓镱瘗癔翊蜴嗌翳一
+yin1	因音阴姻殷茵荫喑湮氤堙洇铟
+yin2	银吟寅淫垠鄞霪狺夤圻龈饮引隐
+yin3	引隐饮瘾殷尹蚓吲
+yin4	印饮荫胤茚窨
+ying1	应英鹰婴樱膺莺罂鹦缨瑛璎撄嘤
+ying2	营迎赢盈蝇莹荧萤萦瀛楹嬴茔滢潆荥蓥影盁
+ying3	影颖颍瘿郢
+ying4	应硬映媵
+yo1	育哟唷
+yo5	哟
+yong1	拥庸佣雍臃邕镛墉慵痈壅鳙饔
+yong2	喁勇永
+yong3	永勇涌踊泳咏俑恿甬蛹
+yong4	用佣
+you1	优幽忧悠攸呦
+you2	由游油邮尤犹柚鱿莸尢铀猷疣蚰蝣蝤繇莜有友
+you3	有友黝酉莠牖铕
+you4	有又右幼诱佑柚囿鼬宥侑蚴釉祐
+yu1	於吁迂淤纡瘀
+yu2	于与余予鱼愚舆娱愉馀逾渔渝俞萸瑜隅揄榆虞禺谀腴竽妤臾欤觎盂窬蝓嵛狳舁雩於雨
+yu3	雨与语予宇羽禹圄屿龉伛圉庾瘐窳俣禺
+yu4	与语育遇狱雨欲预玉愈谷域誉吁蔚寓豫粥郁喻裕浴御驭尉谕毓妪峪芋昱煜熨燠菀蓣饫阈鬻聿钰鹆鹬蜮彧
+yuan1	冤渊鸳眢鸢箢
+yuan2	员元原园源圆缘援袁猿垣辕沅媛芫橼圜塬爰螈鼋远
+yuan3	远
+yuan4	院愿怨苑媛掾垸瑗
+yue1	约曰
+yue4	月越阅跃乐悦岳粤说钥刖瀹栎樾龠钺玥
+yun1	晕氲
+yun2	云匀筠芸耘纭昀郧允
+yun3	允陨殒狁
+yun4	运均韵晕孕蕴酝愠熨郓韫恽
+za1	扎咂匝拶
+za2	杂咱砸
+za3	咋
+zai1	灾哉栽
+zai3	载仔宰崽
+zai4	在再载
+zan1	簪糌
+zan2	咱
+zan3	攒拶昝趱
+zan4	赞暂瓒錾
+zan5	咱
+zang1	赃臧脏
+zang3	驵
+zang4	藏脏葬奘
+zao1	遭糟
+zao2	凿早
+zao3	早澡枣蚤藻
+zao4	造灶躁噪皂燥唣
+ze2	则责泽择咋啧迮帻赜笮箦舴
+ze4	侧仄昃
+zei2	贼
+zen3	怎
+zen4	谮
+zeng1	曾增憎缯罾
+zeng4	赠缯甑锃
+zha1	查扎咋渣喳揸楂哳吒齄
+zha2	炸扎札喋轧闸铡
+zha3	眨砟
+zha4	炸咋诈乍蜡栅榨柞吒咤痄蚱
+zhai1	摘侧斋
+zhai2	宅翟择窄
+zhai3	窄
+zhai4	债祭寨砦瘵
+zhan1	占沾粘瞻詹毡谵旃
+zhan3	展斩辗盏崭搌
+zhan4	战站占颤绽湛蘸栈
+zhang1	张章彰璋蟑樟漳嫜鄣獐
+zhang3	长掌涨仉
+zhang4	丈涨帐障账胀仗杖瘴嶂幛
+zhao1	着招朝嘲昭钊啁
+zhao2	着著
+zhao3	找爪沼
+zhao4	照赵召罩兆肇诏棹笊
+zhe1	折遮蜇
+zhe2	折哲辙辄谪蛰摺磔蜇喆
+zhe3	者褶锗赭
+zhe4	这浙蔗鹧柘
+zhe5	着
+zhei4	这
+zhen1	真针珍斟贞侦甄臻箴砧桢溱蓁椹榛胗祯浈禛
+zhen3	诊枕疹缜畛轸稹
+zhen4	阵镇震圳振赈朕鸩
+zheng1	正争征丁挣症睁徵蒸怔筝铮峥狰钲鲭
+zheng3	整拯
+zheng4	政正证���郑症怔铮诤帧
+zhi1	之只知指支织氏枝汁掷芝吱肢脂蜘栀卮胝祗
+zhi2	直指职值执植殖侄踯摭絷跖埴只纸
+zhi3	只指纸止址旨徵趾咫芷枳祉轵黹酯
+zhi4	知至制识治志致质智置秩滞帜稚挚掷峙窒炙痔栉桎帙轾贽痣豸陟忮彘膣雉鸷骘蛭踬郅觯
+zhong1	中终钟忠衷锺盅忪螽舯
+zhong2	种
+zhong3	种肿踵冢
+zhong4	中种重众仲
+zhou1	周州洲粥舟诌啁
+zhou2	轴妯碡
+zhou3	肘帚
+zhou4	皱骤轴宙咒昼胄纣绉荮籀繇酎
+zhu1	诸朱珠猪株蛛洙诛铢茱邾潴槠橥侏
+zhu2	术逐筑竹烛躅竺舳瘃主
+zhu3	主属煮嘱瞩拄褚渚麈
+zhu4	住注助著驻祝筑柱铸伫贮箸炷蛀杼翥苎疰
+zhua1	抓挝
+zhua3	爪
+zhuai1	拽
+zhuai3	拽
+zhuai4	拽
+zhuan1	专砖颛
+zhuan3	转
+zhuan4	传转赚撰沌篆啭馔
+zhuang1	装庄妆桩
+zhuang3	奘
+zhuang4	状壮撞幢僮戆壵
+zhui1	追锥隹椎骓
+zhui4	坠缀赘惴缒
+zhun1	屯谆肫窀
+zhun3	准
+zhuo1	桌捉卓拙涿焯倬
+zhuo2	着著琢缴灼酌浊濯茁啄斫镯诼禚擢浞卓
+zi1	资咨滋仔姿吱兹孜谘呲龇锱辎淄髭赀孳粢趑觜訾缁鲻嵫
+zi3	子紫仔梓姊籽滓秭笫耔茈訾
+zi4	自字渍恣眦
+zong1	宗踪综棕鬃枞腙
+zong3	总偬
+zong4	纵粽
+zou1	邹诹陬鄹驺鲰
+zou3	走
+zou4	奏揍
+zu1	租菹
+zu2	足族卒镞祖组
+zu3	组祖阻诅俎
+zuan1	钻躜
+zuan3	纂缵
+zuan4	赚钻攥
+zui1	堆
+zui3	嘴咀觜
+zui4	最罪醉蕞
+zun1	尊遵樽鳟
+zun3	撙
+zuo1	作嘬
+zuo2	作昨琢笮左
+zuo3	左佐撮
+zuo4	作做坐座怍胙阼唑祚
+zi5	子
+ge5	个歌
+gu5	顾咕
+biao2	表
+sha2	啥
+er5	儿
+fan5	烦
+tou5	头
+jing2	井
+ji5	记计疾积
+guo5	过
+zhi5	置质制指脂
+niao2	袅
+xi5	息系
+ying5	蝇盈
+jing5	睛
+liang5	量亮粮粱
+mou3	某
+shan2	闪
+lang5	朗
+fen5	分
+xiong5	兄
+bu5	不部
+hou5	候侯
+ben2	本
+zong2	总
+tai5	太
+mou2	谋牟眸缪
+lai5	来
+niang5	娘
+jian2	减柬
+you5	友
+guan2	管馆
+yue5	月
+diao5	掉
+wu5	物务
+xing5	猩
+bing5	病饼
+ruo5	弱
+mu2	亩母
+zuo5	坐做
+chui5	吹
+yun5	晕
+nve4	疟虐
+qie2	茄伽
+zui2	嘴
+rong5	溶
+mei5	妹
+kan5	看
+kao2	考
+he5	合
+sheng5	生
+yang5	鸯扬阳
+cheng5	诚
+lv2	驴榈旅侣
+dai5	袋待
+jin2	紧
+yao5	药
+wang5	望枉王
+jiu2	九
+jun3	菌
+gei2	给
+zheng2	整
+nao5	闹
+gun2	滚
+tong5	同筒
+dian2	点典
+gui2	鬼
+bei2	北
+tao5	桃
+lia2	俩
+yi5	易议一意义
+shen5	神
+ling5	龄
+weng5	嗡
+qi5	气
+qin5	亲
+nai2	奶
+heng5	哼
+nv2	女
+ju5	矩
+ye5	爷
+mian5	面
+rou5	柔
+qia5	恰
+pie2	撇
+guang2	广
+ka2	卡
+ken2	恳
+si2	死
+dang2	党
+yu5	育裕
+wo5	窝
+deng2	等
+fa5	发
+po5	婆
+kong2	孔
+zhen2	诊
+gai2	改
+pin5	频
+wan5	玩
+jiang2	讲奖
+nai5	奶耐
+mu5	目
+shun5	顺
+ne4	讷呐
+dao5	到
+nuan2	暖
+piao5	飘
+su5	诉
+nang5	囊
+long5	咙
+zhu5	住
+pian5	片
+yong5	用
+ou2	偶
+hu5	乎
+zou2	走
+bai5	白
+zhang2	长
+bang2	榜
+zhan2	展
+huo5	活火
+tie2	铁
+du5	度
+zhuan2	转
+chou5	酬
+gan5	杆
+sou5	嗽
+hai5	害
+jiong2	炯
+jiu5	舅
+duo5	朵
+o1	噢喔
+niu1	妞
+sai5	噻
+pou2	锫

download_default_datalist.py

@@ -0,0 +1,104 @@
+# !/usr/bin/env python3
+# -*- coding: utf-8 -*-
+#
+# Copyright 2016-2099 Ailemon.net
+#
+# This file is part of ASRT Speech Recognition Tool.
+#
+# ASRT is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+# ASRT is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with ASRT.  If not, see <https://www.gnu.org/licenses/>.
+# ============================================================================
+
+"""
+@author: nl8590687
+用于下载ASRT语音识别系统声学模型训练默认用的数据集列表程序
+"""
+
+import os
+import logging
+import json
+import requests
+import zipfile
+
+logging.basicConfig(
+    format='%(asctime)s - %(pathname)s[line:%(lineno)d] - %(levelname)s: %(message)s',
+    level=logging.INFO)
+
+DEFAULT_DATALIST_PATH = 'datalist/'
+if not os.path.exists(DEFAULT_DATALIST_PATH):
+    os.makedirs(DEFAULT_DATALIST_PATH)
+
+URL_DATALIST_INDEX = "https://d.ailemon.net/asrt_assets/datalist/index.json"
+rsp_index = requests.get(URL_DATALIST_INDEX)
+rsp_index.encoding = 'utf-8'
+if rsp_index.ok:
+    logging.info('Has connected to ailemon\'s download server...')
+else:
+    logging.error('%s%s', 'Can not connected to ailemon\'s download server.',
+                  'please check your network connection.')
+
+index_json = json.loads(rsp_index.text)
+if index_json['status_code'] != 200:
+    raise Exception(index_json['status_message'])
+
+body = index_json['body']
+logging.info('start to download datalist from ailemon\'s download server...')
+
+url_prefix = body['url_prefix']
+for i in range(len(body['datalist'])):
+    print(i, body['datalist'][i]['name'])
+print(len(body['datalist']), 'all datalist')
+num = input('Please choose which you select: (default all)')
+if len(num) == 0:
+    num = len(body['datalist'])
+else:
+    num = int(num)
+
+
+def deal_download(datalist_item, url_prefix_str, datalist_path):
+    """
+    to deal datalist file download
+    """
+    logging.info('%s%s', 'start to download datalist ', datalist_item['name'])
+    save_path = datalist_path
+    if not os.path.exists(save_path):
+        os.makedirs(save_path)
+        logging.info('%s`%s`', 'Created directory ', save_path)
+
+    zipfilename = datalist_item['zipfile']
+    tmp_url = url_prefix_str + zipfilename
+    save_filename = os.path.join(save_path, zipfilename)
+    rsp = requests.get(tmp_url)
+    with open(save_filename, "wb") as file_pointer:
+        file_pointer.write(rsp.content)
+    if rsp.ok:
+        logging.info('%s `%s` %s', 'Download', zipfilename, 'complete')
+    else:
+        logging.error('%s%s%s%s%s', 'Can not download ', zipfilename,
+                      ' from ailemon\'s download server. ',
+                      'http status ok is ', str(rsp.ok))
+
+    f = zipfile.ZipFile(save_filename, 'r')  # 压缩文件位置
+    f.extractall(save_path)
+    f.close()
+    logging.info('%s `%s` %s', 'unzip', zipfilename, 'complete')
+
+
+if num == len(body['datalist']):
+    for i in range(len(body['datalist'])):
+        deal_download(body['datalist'][i], body['url_prefix'], DEFAULT_DATALIST_PATH)
+else:
+    deal_download(body['datalist'][num], body['url_prefix'], DEFAULT_DATALIST_PATH)
+
+logging.info('%s%s%s', 'Datalist files download complete. ',
+             'Please remember to download these datasets from ',
+             body['dataset_download_page_url'])

epoch_SpeechModel251bn.txt

@@ -0,0 +1 @@
+save_models\SpeechModel251bn\SpeechModel251bn_epoch19

evaluate_speech_model.py

@@ -0,0 +1,50 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+#
+# Copyright 2016-2099 Ailemon.net
+#
+# This file is part of ASRT Speech Recognition Tool.
+#
+# ASRT is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+# ASRT is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with ASRT.  If not, see <https://www.gnu.org/licenses/>.
+# ============================================================================
+
+"""
+@author: nl8590687
+用于测试语音识别系统语音模型的程序
+"""
+
+import os
+
+from speech_model import ModelSpeech
+from model_zoo.speech_model.keras_backend import SpeechModel251BN
+from data_loader import DataLoader
+from speech_features import Spectrogram
+
+os.environ["CUDA_VISIBLE_DEVICES"] = "0"
+
+AUDIO_LENGTH = 1600
+AUDIO_FEATURE_LENGTH = 200
+CHANNELS = 1
+# 默认输出的拼音的表示大小是1428，即1427个拼音+1个空白块
+OUTPUT_SIZE = 1428
+sm251bn = SpeechModel251BN(
+    input_shape=(AUDIO_LENGTH, AUDIO_FEATURE_LENGTH, CHANNELS),
+    output_size=OUTPUT_SIZE
+)
+feat = Spectrogram()
+evalue_data = DataLoader('dev')
+ms = ModelSpeech(sm251bn, feat, max_label_length=64)
+
+ms.load_model('save_models/' + sm251bn.get_model_name() + '.model.h5')
+ms.evaluate_model(data_loader=evalue_data, data_count=-1,
+                  out_report=True, show_ratio=True, show_per_step=100)

language_model3.py

@@ -0,0 +1,123 @@
+# !/usr/bin/env python3
+# -*- coding: utf-8 -*-
+#
+# Copyright 2016-2099 Ailemon.net
+#
+# This file is part of ASRT Speech Recognition Tool.
+#
+# ASRT is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+# ASRT is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with ASRT.  If not, see <https://www.gnu.org/licenses/>.
+# ============================================================================
+
+"""
+@author: nl8590687
+ASRT语音识别的语言模型
+
+基于N-Gram的语言模型
+"""
+
+import os
+
+from utils.ops import get_symbol_dict, get_language_model
+
+
+class ModelLanguage:
+    """
+    ASRT专用N-Gram语言模型
+    """
+
+    def __init__(self, model_path: str):
+        self.model_path = model_path
+        self.dict_pinyin = dict()
+        self.model1 = dict()
+        self.model2 = dict()
+
+    def load_model(self):
+        """
+        加载N-Gram语言模型到内存
+        """
+        self.dict_pinyin = get_symbol_dict('dict.txt')
+        self.model1 = get_language_model(os.path.join(self.model_path, 'language_model1.txt'))
+        self.model2 = get_language_model(os.path.join(self.model_path, 'language_model2.txt'))
+        model = (self.dict_pinyin, self.model1, self.model2)
+        return model
+
+    def pinyin_to_text(self, list_pinyin: list, beam_size: int = 100) -> str:
+        """
+        拼音转文本，一次性取得全部结果
+        """
+        result = list()
+        tmp_result_last = list()
+        for item_pinyin in list_pinyin:
+            tmp_result = self.pinyin_stream_decode(tmp_result_last, item_pinyin, beam_size)
+            if len(tmp_result) == 0 and len(tmp_result_last) > 0:
+                result.append(tmp_result_last[0][0])
+                tmp_result = self.pinyin_stream_decode([], item_pinyin, beam_size)
+                if len(tmp_result) > 0:
+                    result.append(tmp_result[0][0])
+                tmp_result = []
+            tmp_result_last = tmp_result
+
+        if len(tmp_result_last) > 0:
+            result.append(tmp_result_last[0][0])
+
+        return ''.join(result)
+
+    def pinyin_stream_decode(self, temple_result: list,
+                             item_pinyin: str,
+                             beam_size: int = 100) -> list:
+        """
+        拼音流式解码，逐字转换，每次返回中间结果
+        """
+        # 如果这个拼音不在汉语拼音字典里的话，直接返回空列表，不做decode
+        if item_pinyin not in self.dict_pinyin:
+            return []
+
+        # 获取拼音下属的字的列表，cur_words包含了该拼音对应的所有的字
+        cur_words = self.dict_pinyin[item_pinyin]
+        # 第一个字做初始处理
+        if len(temple_result) == 0:
+            lst_result = list()
+            for word in cur_words:
+                # 添加该字到可能的句子列表，设置初始概率为1.0
+                lst_result.append([word, 1.0])
+            return lst_result
+
+        # 开始处理已经至少有一个字的中间结果情况
+        new_result = list()
+        for sequence in temple_result:
+            for cur_word in cur_words:
+                # 得到2-gram的汉字子序列
+                tuple2_word = sequence[0][-1] + cur_word
+                if tuple2_word not in self.model2:
+                    # 如果2-gram子序列不存在
+                    continue
+                # 计算状态转移概率
+                prob_origin = sequence[1]  # 原始概率
+                count_two_word = float(self.model2[tuple2_word])  # 二字频数
+                count_one_word = float(self.model1[tuple2_word[-2]])  # 单字频数
+                cur_probility = prob_origin * count_two_word / count_one_word
+                new_result.append([sequence[0] + cur_word, cur_probility])
+
+        new_result = sorted(new_result, key=lambda x: x[1], reverse=True)
+        if len(new_result) > beam_size:
+            return new_result[0:beam_size]
+        return new_result
+
+
+if __name__ == '__main__':
+    ml = ModelLanguage('model_language')
+    ml.load_model()
+
+    _str_pinyin = ['zhe4', 'zhen1', 'shi4', 'ji2', 'hao3', 'de5']
+    _RESULT = ml.pinyin_to_text(_str_pinyin)
+    print('语音转文字结果:\n', _RESULT)

model_language/language_model1.txt

@@ -0,0 +1,6880 @@
+3941753
+的	213994
+一	72404
+是	69047
+在	63563
+不	61699
+了	60856
+有	59005
+人	52706
+中	52005
+国	50797
+我	48825
+大	45439
+上	39507
+年	37962
+为	35373
+日	34764
+和	34386
+个	33555
+你	32699
+网	31057
+这	30804
+会	30280
+新	30104
+来	29692
+时	29282
+公	28564
+发	28408
+以	28300
+生	27753
+要	27273
+业	27124
+行	26852
+出	26498
+到	26489
+能	25061
+对	24333
+下	23178
+作	23163
+家	23126
+就	22752
+可	22633
+成	22287
+用	22093
+们	21609
+信	21375
+分	21249
+多	21249
+后	21091
+月	20948
+经	20808
+于	20536
+方	20097
+动	20036
+市	20010
+者	19994
+也	19903
+全	19557
+地	19486
+学	19286
+天	18970
+过	18839
+现	18730
+他	18655
+机	18457
+好	18418
+本	18266
+报	18105
+进	17950
+自	17942
+都	17939
+|	17935
+高	17912
+考	17686
+还	17516
+场	17343
+前	17327
+体	17173
+开	17170
+车	17162
+子	16765
+关	16739
+面	16739
+最	16319
+务	16141
+去	16139
+将	16101
+力	15732
+得	15534
+金	15473
+看	15460
+电	15372
+合	15351
+说	15327
+点	15278
+资	15195
+主	15175
+小	15141
+工	15051
+部	14941
+实	14817
+产	14750
+与	14722
+司	14679
+美	14551
+手	14526
+品	14454
+比	14444
+游	14057
+加	14056
+同	14040
+而	13944
+内	13931
+之	13826
+等	13818
+联	13778
+明	13741
+长	13699
+定	13485
+法	13367
+事	13354
+所	13284
+外	13221
+理	13216
+心	13150
+通	13094
+区	13081
+文	12982
+球	12976
+备	12972
+赛	12781
+其	12732
+里	12593
+台	12491
+没	12463
+重	12370
+那	12343
+表	12286
+道	12273
+化	12249
+名	12236
+军	12147
+期	12100
+安	12079
+目	12054
+间	11882
+海	11878
+准	11769
+试	11729
+更	11725
+第	11653
+提	11624
+么	11623
+入	11588
+万	11519
+员	11509
+次	11448
+如	11405
+民	11400
+数	11353
+系	11349
+度	11080
+院	10887
+展	10865
+建	10850
+三	10848
+元	10808
+位	10720
+保	10648
+政	10645
+平	10633
+计	10538
+服	10500
+相	10466
+起	10399
+限	10343
+意	10213
+及	10148
+性	10049
+广	10041
+两	10021
+情	10013
+当	9828
+号	9805
+女	9802
+传	9801
+息	9782
+线	9724
+从	9714
+制	9665
+北	9578
+利	9571
+题	9556
+交	9556
+任	9510
+但	9471
+京	9438
+特	9422
+二	9378
+证	9330
+打	9320
+队	9313
+很	9282
+西	9218
+着	9217
+已	9209
+然	9193
+量	9191
+正	9172
+或	9046
+据	9041
+此	9017
+版	8996
+并	8969
+视	8946
+老	8925
+战	8923
+科	8909
+价	8898
+给	8861
+设	8737
+黄	8733
+今	8657
+示	8566
+水	8560
+被	8552
+站	8549
+无	8522
+许	8468
+式	8438
+总	8416
+果	8384
+记	8343
+单	8342
+权	8297
+东	8284
+推	8258
+专	8242
+商	8237
+因	8203
+身	8163
+强	8158
+约	8129
+应	8116
+想	8075
+图	8070
+问	7972
+省	7948
+影	7893
+路	7884
+代	7858
+投	7837
+让	7833
+基	7770
+管	7766
+导	7726
+戏	7705
+托	7679
+做	7679
+南	7668
+儿	7627
+技	7613
+回	7602
+群	7556
+活	7544
+样	7516
+布	7454
+友	7447
+华	7413
+解	7396
+收	7390
+接	7378
+告	7360
+件	7343
+型	7341
+首	7269
+知	7246
+注	7218
+卡	7218
+些	7215
+消	7175
+创	7168
+玩	7168
+参	7162
+马	7152
+持	7149
+际	7104
+级	7079
+师	7070
+增	7069
+物	7059
+房	7049
+选	7039
+门	6997
+格	6970
+种	6965
+近	6926
+直	6892
+乐	6857
+责	6852
+股	6847
+别	6831
+调	6829
+英	6813
+户	6780
+城	6768
+教	6742
+向	6726
+标	6701
+运	6681
+各	6676
+育	6667
+达	6656
+程	6641
+至	6622
+费	6600
+局	6585
+尔	6568
+常	6562
+社	6560
+色	6516
+话	6472
+斯	6467
+受	6434
+周	6430
+原	6417
+山	6381
+立	6379
+微	6368
+结	6365
+转	6330
+世	6323
+界	6304
+项	6287
+少	6227
+造	6211
+统	6208
+医	6202
+感	6201
+只	6189
+领	6183
+片	6162
+张	6162
+需	6152
+啊	6129
+款	6114
+源	6101
+均	6065
+气	6062
+带	6059
+指	6047
+买	6043
+己	6032
+风	6018
+真	6007
+认	5981
+购	5975
+研	5973
+雪	5944
+论	5921
+放	5885
+节	5867
+复	5846
+超	5846
+查	5825
+取	5776
+规	5761
+容	5734
+该	5715
+十	5714
+空	5713
+江	5711
+求	5684
+她	5675
+头	5656
+由	5649
+再	5636
+观	5622
+术	5610
+完	5606
+招	5590
+快	5585
+亚	5582
+深	5553
+职	5551
+组	5543
+请	5524
+把	5512
+热	5481
+处	5461
+营	5455
+星	5444
+流	5436
+未	5433
+改	5430
+演	5427
+办	5418
+先	5413
+怀	5412
+类	5378
+州	5378
+质	5369
+变	5362
+口	5355
+吗	5346
+易	5289
+使	5270
+才	5259
+预	5249
+排	5236
+客	5223
+永	5221
+四	5211
+续	5201
+供	5184
+闻	5177
+济	5177
+共	5175
+众	5165
+爱	5159
+男	5158
+支	5157
+整	5157
+配	5131
+精	5107
+播	5088
+光	5074
+讯	5065
+企	5058
+称	4990
+每	4967
+非	4940
+宣	4929
+圈	4925
+什	4887
+王	4885
+值	4861
+速	4840
+太	4832
+互	4812
+红	4758
+包	4749
+升	4739
+何	4738
+集	4737
+亿	4736
+份	4736
+德	4722
+校	4718
+确	4705
+优	4702
+团	4693
+宝	4688
+吧	4656
+几	4638
+足	4633
+云	4627
+媒	4611
+装	4599
+博	4598
+拉	4566
+娃	4557
+远	4551
+又	4548
+议	4523
+击	4502
+财	4470
+案	4462
+允	4459
+融	4418
+条	4414
+翁	4409
+始	4385
+显	4359
+见	4336
+词	4334
+李	4314
+克	4309
+载	4305
+获	4281
+编	4277
+声	4273
+步	4270
+决	4262
+花	4248
+环	4240
+势	4238
+较	4236
+率	4232
+牌	4226
+具	4219
+难	4148
+���	4131
+评	4130
+银	4128
+功	4122
+住	4114
+愿	4114
+置	4087
+效	4086
+神	4068
+剧	4056
+录	4051
+富	4042
+究	4037
+防	4012
+官	4004
+林	4001
+引	3999
+举	3991
+细	3988
+书	3986
+警	3980
+走	3961
+兄	3960
+构	3942
+五	3937
+反	3935
+低	3926
+态	3919
+享	3909
+连	3895
+春	3893
+维	3888
+普	3875
+治	3844
+白	3831
+照	3826
+销	3815
+险	3810
+策	3781
+季	3771
+像	3769
+听	3763
+助	3760
+况	3754
+士	3734
+验	3733
+望	3726
+器	3724
+且	3721
+划	3710
+清	3700
+边	3700
+监	3683
+识	3670
+形	3654
+吃	3636
+百	3632
+候	3618
+跟	3618
+佳	3600
+习	3598
+频	3594
+随	3578
+络	3562
+护	3549
+模	3531
+晚	3521
+施	3499
+觉	3491
+测	3455
+仅	3447
+呢	3443
+段	3436
+域	3413
+越	3408
+则	3405
+字	3400
+即	3391
+奖	3386
+料	3386
+航	3385
+历	3354
+健	3352
+湾	3347
+龙	3347
+党	3342
+盖	3342
+双	3337
+福	3322
+汽	3304
+律	3275
+养	3275
+宁	3275
+朋	3255
+售	3254
+协	3238
+米	3222
+它	3214
+范	3209
+音	3209
+读	3208
+青	3195
+核	3184
+欢	3165
+府	3151
+失	3142
+河	3140
+采	3134
+算	3133
+页	3120
+阳	3114
+奥	3113
+弟	3111
+负	3110
+智	3097
+班	3093
+存	3082
+罗	3074
+控	3050
+火	3048
+响	3045
+言	3031
+苏	3030
+拥	3027
+致	3025
+找	3022
+落	3010
+除	2997
+某	2996
+列	2987
+哪	2976
+简	2974
+钱	2972
+终	2970
+根	2967
+食	2959
+港	2954
+巴	2953
+语	2952
+益	2947
+刚	2946
+怎	2937
+紧	2931
+念	2926
+勇	2923
+干	2918
+旅	2908
+检	2901
+介	2900
+景	2894
+极	2893
+早	2886
+争	2868
+拍	2855
+满	2832
+座	2831
+练	2828
+歌	2825
+依	2820
+疑	2798
+雄	2792
+离	2790
+思	2786
+喜	2778
+村	2769
+底	2768
+韩	2763
+必	2762
+登	2747
+缺	2747
+志	2740
+盘	2736
+胸	2736
+素	2733
+象	2723
+训	2721
+帮	2715
+竞	2712
+角	2712
+积	2709
+板	2707
+付	2706
+免	2703
+希	2703
+境	2690
+聘	2686
+冠	2682
+农	2678
+属	2678
+半	2678
+端	2669
+波	2649
+眼	2639
+详	2633
+飞	2627
+店	2621
+妈	2616
+破	2614
+卫	2599
+孩	2595
+稳	2590
+园	2589
+康	2581
+断	2580
+陆	2579
+审	2578
+往	2575
+便	2575
+阿	2575
+陈	2569
+绩	2568
+初	2564
+降	2561
+按	2555
+轻	2541
+够	2541
+床	2539
+材	2538
+居	2533
+却	2531
+继	2529
+套	2518
+刘	2512
+县	2506
+君	2501
+甲	2498
+岁	2495
+减	2480
+扫	2477
+黑	2472
+欧	2467
+亲	2464
+死	2463
+拿	2461
+严	2456
+香	2448
+访	2437
+浪	2434
+批	2430
+适	2426
+送	2424
+独	2418
+例	2416
+码	2404
+石	2402
+违	2393
+讲	2389
+胜	2387
+婚	2375
+病	2370
+纪	2369
+充	2366
+择	2345
+辞	2337
+曾	2336
+留	2336
+补	2336
+略	2334
+章	2329
+围	2326
+命	2321
+攻	2316
+彩	2309
+层	2307
+票	2306
+叫	2304
+届	2302
+阅	2299
+移	2298
+承	2293
+轮	2286
+签	2284
+辑	2282
+义	2282
+换	2279
+艺	2258
+哈	2254
+熊	2252
+含	2249
+占	2249
+奇	2249
+否	2246
+兰	2228
+母	2224
+千	2222
+短	2208
+突	2204
+担	2189
+货	2180
+尼	2176
+额	2175
+杨	2175
+左	2170
+铁	2170
+史	2164
+丽	2162
+迎	2152
+古	2151
+课	2149
+牙	2136
+荐	2135
+午	2131
+油	2129
+腾	2128
+滑	2128
+血	2127
+皮	2127
+孕	2126
+察	2114
+询	2110
+谈	2095
+副	2094
+穿	2091
+钟	2084
+威	2084
+泳	2083
+币	2082
+错	2081
+待	2077
+另	2076
+湖	2074
+洲	2068
+爆	2067
+药	2063
+衣	2063
+尽	2058
+良	2058
+背	2058
+哥	2056
+楼	2049
+武	2049
+印	2048
+兴	2044
+索	2037
+税	2027
+密	2020
+临	2020
+坛	2017
+松	2017
+您	2017
+写	2016
+止	2001
+键	2000
+诉	1999
+画	1996
+状	1992
+执	1989
+耍	1987
+姐	1986
+冰	1983
+切	1981
+雅	1978
+亮	1977
+余	1963
+睡	1959
+停	1958
+温	1952
+粉	1949
+差	1947
+八	1944
+雷	1941
+跑	1932
+织	1931
+室	1931
+申	1928
+启	1923
+涨	1921
+坚	1921
+综	1916
+丝	1916
+庆	1901
+括	1900
+夫	1898
+酒	1891
+追	1889
+久	1889
+笔	1888
+假	1886
+虽	1883
+革	1882
+秀	1881
+岛	1876
+弹	1867
+础	1864
+土	1864
+搭	1861
+汇	1861
+培	1855
+析	1846
+幅	1845
+答	1838
+邮	1830
+皇	1824
+朗	1817
+族	1817
+故	1800
+伤	1796
+善	1782
+须	1774
+席	1770
+压	1768
+授	1767
+券	1766
+贸	1763
+盟	1757
+镇	1757
+退	1757
+谁	1749
+迷	1736
+右	1725
+绝	1723
+露	1722
+藏	1719
+鱼	1718
+修	1717
+贷	1709
+妹	1708
+订	1706
+激	1699
+似	1698
+街	1698
+迪	1693
+贵	1691
+纳	1684
+陪	1681
+沙	1673
+卖	1672
+遇	1661
+挑	1658
+川	1657
+央	1651
+六	1637
+射	1631
+探	1627
+仍	1617
+舰	1612
+悉	1601
+礼	1599
+梦	1592
+丰	1579
+娱	1575
+吸	1561
+曲	1559
+疗	1555
+夏	1549
+牛	1546
+摄	1543
+述	1538
+库	1534
+损	1531
+尚	1530
+篮	1524
+杀	1522
+令	1521
+裙	1519
+惠	1509
+庭	1508
+箱	1508
+津	1507
+婆	1506
+刷	1503
+竟	1501
+木	1501
+啦	1494
+脑	1492
+绍	1488
+障	1485
+典	1479
+穷	1479
+榜	1478
+禁	1478
+输	1476
+靠	1476
+荣	1476
+潜	1472
+毕	1471
+署	1462
+透	1462
+梅	1456
+判	1455
+倍	1446
+派	1442
+峰	1438
+困	1437
+绿	1436
+娟	1430
+秘	1427
+伦	1426
+谷	1423
+灵	1422
+乡	1417
+诺	1414
+巨	1409
+幕	1403
+味	1402
+跌	1401
+鸡	1401
+七	1397
+谢	1395
+鲁	1394
+摸	1385
+雨	1384
+蛋	1381
+夜	1378
+熟	1376
+森	1375
+涉	1371
+顾	1368
+泰	1366
+册	1358
+镜	1355
+昨	1353
+���	1350
+脸	1350
+苹	1343
+萨	1343
+暴	1331
+附	1330
+毛	1328
+链	1327
+顺	1327
+馆	1327
+抱	1326
+唱	1323
+澳	1318
+督	1318
+贝	1317
+笑	1316
+扩	1314
+倒	1314
+厅	1313
+蓝	1312
+害	1311
+哦	1310
+舞	1307
+守	1306
+寻	1306
+田	1304
+匀	1302
+逐	1299
+兵	1297
+聚	1296
+救	1292
+汉	1287
+辆	1286
+征	1279
+阵	1270
+浙	1268
+洋	1268
+甚	1267
+菜	1263
+阶	1262
+恐	1261
+概	1259
+滚	1258
+驾	1256
+趣	1255
+促	1253
+块	1251
+厂	1246
+卷	1236
+赵	1234
+肉	1233
+览	1232
+魔	1232
+盛	1228
+乖	1228
+顶	1227
+搜	1223
+乎	1221
+冲	1221
+抓	1221
+途	1219
+诚	1217
+侧	1215
+旗	1209
+润	1207
+归	1202
+袭	1197
+针	1195
+狂	1195
+童	1193
+脚	1190
+贴	1188
+延	1186
+惊	1186
+挂	1186
+架	1186
+岸	1185
+异	1185
+隐	1183
+序	1180
+毒	1176
+刻	1173
+伴	1172
+操	1165
+嘛	1161
+般	1160
+肥	1159
+赶	1159
+朝	1154
+耳	1153
+休	1151
+呼	1149
+伟	1149
+估	1148
+颜	1147
+稿	1145
+杯	1145
+饭	1141
+晓	1135
+斗	1133
+俄	1124
+啥	1122
+翻	1119
+予	1118
+急	1118
+抗	1118
+呀	1114
+鲜	1113
+坐	1110
+塞	1109
+妇	1108
+剂	1104
+轩	1098
+屏	1095
+九	1094
+豪	1092
+献	1091
+掉	1091
+吴	1090
+野	1087
+恩	1087
+尤	1086
+挥	1084
+瑞	1075
+疆	1070
+旧	1070
+曼	1067
+址	1065
+篇	1063
+遭	1058
+赞	1058
+聊	1057
+静	1053
+距	1050
+句	1050
+跳	1049
+净	1047
+硬	1044
+剑	1043
+犯	1042
+截	1040
+若	1040
+努	1038
+伙	1036
+饰	1031
+抢	1029
+软	1028
+幻	1021
+淮	1018
+捐	1017
+肯	1015
+描	1014
+慢	1014
+齐	1012
+束	1006
+船	1001
+辽	1000
+凡	999
+父	999
+赢	996
+败	994
+嘉	994
+虑	993
+脱	993
+帅	993
+累	991
+幸	989
+塔	986
+援	984
+封	982
+徐	982
+怕	980
+符	976
+孙	975
+梁	972
+扬	970
+敏	967
+裁	966
+零	966
+吉	963
+草	961
+措	959
+贫	959
+龄	957
+痛	957
+纸	950
+洗	946
+曝	945
+患	944
+映	940
+筑	938
+危	934
+讨	934
+赏	933
+猫	932
+冤	932
+罪	927
+堂	924
+窗	922
+沉	922
+丹	921
+凤	921
+励	917
+咨	913
+偏	911
+宽	911
+亡	910
+罚	909
+敢	908
+召	907
+拟	903
+纷	896
+伊	895
+莱	891
+胡	891
+递	881
+蒙	881
+乱	881
+驶	880
+嗡	878
+肤	876
+震	869
+兽	867
+冷	866
+圳	863
+蒂	861
+刺	861
+磁	860
+锋	859
+夺	858
+巡	858
+著	857
+释	854
+惹	849
+储	845
+肌	838
+杰	837
+厦	835
+奶	833
+固	832
+宗	831
+账	829
+琼	829
+烈	827
+恋	827
+霞	826
+渐	824
+宏	824
+仪	819
+暗	818
+赚	817
+桥	816
+沃	815
+跃	815
+纯	815
+捷	814
+侠	814
+跨	813
+敦	813
+慈	813
+档	812
+醒	811
+艇	809
+餐	809
+舒	809
+坏	809
+胎	808
+丁	807
+遗	805
+怪	803
+喝	803
+趋	802
+触	800
+避	799
+莫	799
+私	797
+弱	796
+焦	795
+沿	795
+掌	795
+劳	789
+辣	787
+折	786
+慧	786
+凶	785
+宜	782
+握	780
+覆	780
+鼓	779
+末	778
+摩	778
+苦	778
+爸	777
+租	776
+宇	775
+潮	774
+郑	773
+树	772
+刊	770
+玉	770
+弃	769
+迅	768
+骗	765
+忘	765
+蔡	763
+锦	762
+枉	759
+歪	757
+嫌	756
+灯	756
+刀	752
+邀	751
+既	751
+暖	751
+污	749
+桃	748
+误	748
+搞	748
+宅	746
+吻	745
+夹	740
+硕	734
+勒	734
+俊	733
+饮	732
+遍	732
+戴	731
+凯	731
+鞋	730
+酷	729
+洛	727
+宫	726
+债	725
+暂	725
+耀	725
+秒	724
+沈	722
+拜	722
+董	721
+漫	721
+萌	721
+扶	720
+染	720
+恒	718
+爽	717
+劲	715
+唯	715
+虎	713
+腿	712
+狗	711
+懂	710
+冒	709
+辅	708
+钢	708
+散	703
+岗	702
+菲	702
+欣	700
+唐	700
+撞	699
+煤	699
+驱	698
+寸	694
+甩	693
+杂	692
+甘	691
+尊	689
+猪	689
+纽	688
+摔	687
+呈	685
+横	685
+籍	685
+晨	685
+扯	685
+顿	684
+池	682
+渠	680
+祝	680
+枪	678
+锁	677
+麻	677
+杆	676
+凭	676
+庄	676
+抽	675
+姆	670
+衡	670
+宾	668
+偶	665
+秦	665
+畅	664
+忙	664
+疾	663
+坦	662
+缘	658
+驻	656
+茨	656
+仁	655
+豆	652
+虚	650
+屋	650
+泽	649
+厚	647
+混	647
+兼	646
+拐	640
+拆	639
+洁	639
+替	638
+残	638
+艾	637
+甜	636
+填	635
+纺	634
+圆	632
+瓷	631
+挺	630
+杭	629
+募	628
+糖	628
+壁	625
+瓦	622
+茶	620
+骑	619
+浏	618
+仙	618
+默	613
+缓	612
+麦	612
+幼	611
+阴	611
+妻	610
+泛	609
+楚	609
+俱	607
+扣	607
+珠	606
+仔	606
+骨	605
+辉	605
+绕	604
+忍	604
+郭	604
+拼	602
+卓	599
+琳	597
+晋	595
+冬	595
+敌	594
+恶	593
+症	591
+瓜	589
+沟	588
+妆	586
+携	585
+筹	585
+奔	585
+磊	583
+羊	583
+缩	582
+贯	582
+淘	581
+谓	578
+巧	578
+圣	578
+帝	578
+紫	577
+烟	577
+盗	575
+羽	575
+洪	575
+佛	573
+娜	573
+逼	572
+彻	571
+赴	570
+炫	570
+返	569
+摆	569
+姓	568
+矿	567
+猛	565
+尾	565
+叶	565
+朱	563
+娘	563
+徽	562
+酸	561
+役	560
+撑	560
+哭	560
+拓	559
+霸	559
+炮	559
+淡	558
+悬	557
+勤	557
+乘	556
+揭	556
+剩	554
+坡	551
+佩	550
+隔	549
+乔	548
+烧	548
+刑	546
+挖	545
+诗	545
+乌	542
+闲	542
+宋	540
+颖	538
+逆	537
+阻	533
+眠	531
+扎	530
+闭	529
+忠	529
+塑	528
+迹	528
+秋	524
+殊	521
+艳	521
+葛	521
+伍	518
+疯	518
+缴	517
+忆	517
+锐	515
+珍	515
+怖	514
+侵	512
+陕	511
+墨	511
+乏	511
+抵	507
+倾	507
+凰	506
+拔	505
+匹	504
+液	504
+杜	503
+泉	501
+拒	500
+毫	498
+逻	498
+涌	495
+媛	494
+燃	493
+繁	493
+碰	493
+添	492
+欲	490
+誉	489
+植	488
+迁	484
+寒	484
+耗	484
+桂	484
+鬼	484
+框	483
+伯	483
+胃	481
+炸	481
+黎	481
+押	480
+洞	479
+嘴	478
+奋	476
+渔	476
+墙	476
+唤	475
+陷	475
+惨	474
+径	473
+宿	473
+弄	473
+践	472
+晒	472
+舍	472
+摘	470
+奏	470
+劝	470
+呵	470
+湿	469
+瓶	469
+旋	469
+峡	469
+鹏	468
+吐	466
+敬	466
+肃	465
+贡	465
+隆	464
+偿	462
+爷	462
+衰	459
+赖	457
+耐	455
+惯	453
+偷	453
+振	452
+赔	452
+尝	451
+愉	451
+沪	450
+轨	450
+彭	450
+荡	449
+孔	449
+乳	448
+迈	447
+涯	447
+诊	445
+旬	445
+鳌	442
+玛	442
+昌	442
+祥	441
+凌	440
+寿	439
+诸	439
+裤	439
+浮	437
+滨	435
+喊	435
+柱	434
+滴	434
+旦	433
+鉴	433
+咱	433
+苑	433
+稍	432
+昆	432
+逊	431
+弗	430
+俗	430
+栏	430
+胖	430
+挡	429
+悦	429
+胞	428
+怨	428
+睛	426
+圾	423
+冯	423
+垃	422
+贺	421
+芳	420
+轿	418
+疫	418
+吨	417
+闹	417
+俩	417
+庸	414
+渴	414
+壮	413
+煌	413
+尺	412
+撒	412
+柔	412
+盈	410
+扰	410
+盾	409
+炉	408
+脂	407
+浦	407
+瘦	407
+磨	407
+亦	406
+帕	404
+斤	404
+穆	403
+汤	403
+旁	403
+浓	402
+貌	401
+肖	398
+燕	398
+涛	398
+琴	398
+辛	397
+陶	397
+涵	395
+柳	395
+拦	394
+番	393
+烦	393
+掘	390
+踪	389
+尖	389
+剪	389
+驰	388
+捕	388
+扮	387
+傅	387
+豹	386
+哲	386
+梯	386
+惜	386
+亏	385
+咋	383
+幂	382
+炼	382
+溪	382
+薄	381
+踢	381
+泡	381
+拖	380
+削	380
+祖	379
+嗯	379
+脏	376
+卢	376
+迟	376
+桌	376
+雾	375
+循	375
+辖	374
+埃	373
+邦	373
+薪	372
+昂	372
+披	371
+蒋	371
+晶	371
+渊	371
+艘	369
+魅	369
+蜜	369
+袖	368
+婷	368
+绪	367
+翘	366
+毁	366
+兹	364
+酱	364
+芯	363
+踏	363
+纹	360
+恢	359
+械	359
+粮	359
+姿	358
+翔	358
+卦	357
+姚	357
+胶	357
+奉	355
+婴	354
+臀	352
+苗	352
+诀	350
+纲	350
+鸣	350
+谋	349
+逃	349
+胆	348
+灭	348
+霍	347
+胁	346
+憾	344
+桑	342
+媳	342
+熏	342
+戈	341
+浩	341
+猎	340
+忧	340
+垫	340
+拨	338
+孤	337
+摇	335
+矩	334
+柴	334
+扑	334
+拳	334
+逸	334
+辈	334
+抹	334
+廉	333
+鹿	332
+侦	331
+歉	331
+蹲	331
+粤	330
+腹	330
+铺	330
+扭	329
+骂	329
+伸	328
+牢	327
+翼	327
+爪	327
+荒	326
+玲	326
+腐	325
+仓	325
+寨	324
+咖	323
+漏	323
+插	323
+眉	323
+仿	322
+译	322
+契	322
+泥	322
+兑	321
+牵	320
+喷	319
+醉	319
+肝	318
+腰	318
+傻	318
+薛	318
+纠	317
+肩	317
+棒	316
+辰	315
+朵	315
+呗	315
+姻	313
+糕	311
+劣	309
+闽	309
+忽	308
+诈	307
+凉	307
+柯	306
+旺	306
+岩	306
+氧	305
+膜	305
+躺	305
+毅	302
+鸿	302
+遵	301
+鸟	301
+吓	300
+骅	300
+骷	299
+髅	299
+噢	299
+癌	298
+魂	298
+槽	298
+潘	297
+漂	297
+炒	296
+袋	296
+盒	295
+灰	295
+馈	294
+撕	294
+贾	294
+斌	294
+狮	293
+伪	293
+戒	293
+井	292
+狼	292
+δ	291
+?	291
+耶	291
+莲	291
+廷	290
+丢	290
+翎	289
+吊	289
+欠	289
+?	288
+碑	288
+颗	288
+?	287
+?	287
+?	287
+寄	287
+琪	287
+勋	287
+帐	286
+丸	286
+废	286
+疼	286
+箭	285
+瞬	285
+姑	285
+晖	284
+谱	284
+蓉	284
+坑	284
+蕾	283
+稀	283
+猜	283
+?	282
+?	282
+?	282
+樱	282
+郎	282
+疏	281
+轴	281
+?	280
+?	280
+怒	280
+铭	279
+吹	279
+?	278
+莞	278
+乒	277
+慎	277
+颇	276
+赠	276
+玻	275
+椅	274
+裂	272
+锅	272
+碧	272
+奠	271
+厉	271
+芒	271
+萱	271
+凄	269
+赫	269
+颁	269
+御	269
+尘	269
+脉	269
+聪	268
+鑫	268
+谣	267
+绑	267
+齿	267
+狐	266
+弯	266
+镑	265
+纤	265
+闪	265
+搬	265
+毙	264
+鼎	264
+肚	264
+彼	263
+璃	262
+廊	261
+碎	261
+堆	261
+涂	261
+履	260
+柜	260
+钻	260
+宴	259
+灾	259
+妙	259
+乃	259
+逛	259
+昭	258
+?	257
+愈	257
+蒸	257
+鼻	257
+×	256
+?	256
+?	256
+倡	256
+迫	256
+姜	256
+叔	256
+衔	255
+玄	255
+茂	255
+寡	255
+?	254
+牧	254
+堪	254
+挤	254
+咯	254
+?	253
+贩	253
+氛	253
+墓	253
+袁	253
+嫁	253
+?	252
+旨	252
+矛	252
+荷	252
+?	251
+碍	251
+贤	251
+郁	251
+脾	251
+删	251
+浅	250
+滋	250
+炭	250
+?	249
+?	249
+?	249
+徒	248
+竹	248
+莎	248
+?	247
+?	247
+坊	247
+擦	247
+?	246
+肺	246
+堡	246
+壳	246
+莉	246
+蛮	246
+?	245
+?	245
+锡	245
+盐	245
+莹	245
+寝	245
+?	244
+?	244
+尬	244
+串	244
+?	243
+尴	243
+虫	243
+卵	243
+烂	243
+傲	243
+绘	242
+铜	242
+尿	242
+?	241
+冀	241
+晰	241
+轰	241
+肠	241
+阔	240
+汗	240
+溜	240
+?	239
+卧	239
+芝	238
+纱	238
+÷	237
+?	237
+逾	237
+伏	237
+恰	237
+函	237
+乙	237
+杠	236
+?	235
+?	235
+骏	235
+滩	235
+．	234
+?	234
+耕	234
+汰	234
+犹	234
+斑	234
+疲	234
+饱	233
+晴	232
+?	231
+?	231
+Э	229
+?	229
+曹	229
+?	228
+辨	227
+泪	227
+粗	227
+剖	226
+魏	226
+湘	226
+肿	226
+邓	226
+腔	226
+窃	225
+艰	225
+丈	225
+赤	225
+?	224
+蔬	224
+卜	224
+袜	224
+锻	224
+膝	224
+卿	223
+抖	223
+渡	223
+庞	223
+?	222
+?	221
+?	221
+?	221
+滤	221
+诱	221
+勾	221
+勿	220
+坎	220
+娇	220
+?	218
+辩	218
+孟	217
+á	216
+妨	216
+纵	216
+５	215
+ж	215
+绰	215
+骁	215
+舟	215
+琦	215
+棋	215
+悲	215
+?	214
+捍	214
+棬	214
+伞	214
+妖	214
+?	213
+渝	213
+盲	213
+栋	213
+踩	213
+夸	213
+裸	213
+窝	213
+哎	213
+?	212
+?	212
+抑	211
+奢	211
+侣	211
+汪	211
+砖	211
+胀	210
+岳	210
+萝	210
+遥	209
+炎	209
+?	208
+?	208
+?	208
+惑	208
+垂	208
+缆	207
+裹	207
+懒	207
+暨	206
+雕	206
+吵	206
+粒	205
+泄	205
+呆	205
+?	204
+捧	204
+捞	204
+侨	204
+柏	204
+馨	204
+缝	204
+妮	204
+爬	204
+厌	204
+?	203
+～	203
+?	203
+凑	203
+瑜	203
+瓮	203
+?	202
+?	202
+绎	202
+悄	202
+霜	202
+?	201
+?	201
+６	201
+洽	201
+殖	201
+碳	201
+?	200
+豫	200
+撤	200
+冻	200
+?	199
+?	199
+?	199
+谨	199
+绣	199
+姨	199
+?	198
+?	198
+?	198
+匠	198
+盆	198
+?	197
+?	197
+?	197
+绸	197
+煮	197
+帽	197
+奈	196
+骚	196
+陌	196
+浴	196
+芽	196
+?	195
+珀	195
+碗	195
+慕	195
+抄	195
+3	194
+?	194
+霆	194
+?	193
+秩	193
+枚	193
+寺	193
+铠	193
+п	192
+д	192
+з	192
+溢	192
+伐	192
+铃	192
+剥	192
+鸳	192
+彰	191
+梳	191
+堵	191
+斜	191
+旭	191
+к	190
+诞	190
+?	189
+叹	189
+凝	189
+飬	188
+?	188
+罩	188
+欺	188
+?	187
+?	187
+霾	187
+撼	187
+臂	187
+渤	187
+?	186
+?	186
+?	186
+乓	186
+抛	186
+?	185
+?	185
+?	185
+?	185
+菌	185
+鹰	185
+坤	185
+逗	185
+鸯	185
+狱	184
+赋	184
+飘	184
+哇	184
+?	183
+?	183
+亩	183
+枣	183
+闯	183
+腻	183
+棉	182
+怡	182
+?	181
+?	181
+?	181
+凸	181
+淀	181
+厨	181
+谊	181
+吕	181
+?	180
+邻	180
+仇	180
+?	179
+?	179
+岭	179
+鼠	179
+晕	179
+?	178
+?	178
+?	178
+骤	178
+擅	178
+?	177
+拘	177
+?	176
+衍	176
+钙	176
+葡	176
+?	175
+沧	175
+宠	175
+蜂	175
+?	174
+▲	174
+贪	174
+巅	173
+爵	173
+窄	173
+塘	173
+恨	173
+４	172
+萄	172
+慰	172
+抚	172
+钧	172
+辕	172
+?	171
+剃	171
+卸	171
+蓬	171
+仲	171
+椒	171
+?	170
+腕	170
+鹅	170
+?	169
+涅	169
+叙	169
+耿	169
+●	168
+?	168
+?	168
+屡	168
+叉	168
+悠	168
+侯	168
+в	167
+?	167
+愤	167
+攀	167
+抬	167
+刮	167
+妥	167
+厢	167
+靴	167
+阎	166
+逢	166
+尹	166
+昏	166
+?	165
+舆	165
+掀	165
+棚	165
+忌	165
+皆	165
+?	164
+?	164
+?	164
+敖	164
+仰	164
+屈	164
+?	163
+?	163
+?	163
+狠	163
+巩	162
+薇	162
+８	161
+槛	161
+瘤	161
+滥	161
+咪	161
+舌	161
+?	160
+９	160
+弊	160
+烤	160
+暑	160
+?	159
+?	159
+宪	159
+佐	159
+泊	159
+钓	159
+ó	158
+?	158
+娴	158
+睿	158
+丑	158
+衫	158
+?	157
+γ	157
+м	157
+蜀	157
+翰	157
+囊	157
+绵	157
+瑟	157
+垮	157
+?	156
+㎡	156
+?	156
+弥	156
+?	155
+擎	155
+詹	155
+劫	155
+珊	155
+陵	155
+悔	155
+茅	154
+唇	154
+鄂	154
+幽	154
+?	153
+?	153
+曦	153
+萍	153
+俺	153
+?	152
+彗	152
+祸	152
+?	151
+?	151
+钞	151
+坞	151
+敲	151
+俏	151
+仑	151
+伽	151
+帆	151
+菠	151
+?	150
+睐	150
+砍	150
+尸	150
+昔	150
+咏	150
+?	149
+?	149
+酬	149
+丧	149
+谐	149
+啡	149
+夕	149
+澡	149
+?	148
+涡	148
+挽	148
+炅	148
+虹	148
+鸭	148
+肾	148
+咬	148
+泼	148
+「	147
+」	147
+Υ	147
+?	147
+吁	147
+歼	147
+橘	147
+?	146
+κ	146
+и	146
+顣	146
+璧	146
+崔	146
+孝	146
+淑	146
+眷	146
+н	145
+?	145
+б	145
+谦	145
+叠	145
+兆	145
+勃	145
+嫩	145
+猴	145
+巾	145
+?	144
+?	144
+?	144
+弘	144
+僵	144
+阁	144
+叵	144
+?	143
+?	143
+?	143
+?	143
+筛	143
+铝	143
+祭	143
+笼	143
+膏	143
+７	142
+∶	142
+?	142
+蚁	142
+曙	142
+蓄	142
+氏	142
+邱	142
+蠢	142
+菱	141
+讼	141
+汁	141
+俑	141
+?	140
+肪	140
+韦	140
+挣	140
+亭	140
+谭	140
+瑶	140
+?	139
+?	139
+Χ	139
+秉	139
+蚂	139
+惩	139
+掏	139
+灿	139
+?	138
+?	138
+?	138
+迄	138
+?	138
+陇	138
+纬	138
+螺	138
+崖	138
+怜	138
+颠	137
+辐	137
+禽	137
+?	136
+?	136
+薯	136
+寓	136
+慌	136
+桶	136
+帖	136
+莓	136
+瑕	136
+?	135
+①	135
+?	135
+?	134
+澎	134
+鹤	134
+廖	134
+饼	134
+?	133
+μ	133
+ε	133
+衷	133
+裕	133
+竭	133
+兔	133
+闷	133
+?	132
+→	132
+ο	132
+谍	132
+祺	132
+崇	132
+搏	132
+愁	132
+掩	132
+屁	132
+ú	131
+堰	131
+潇	131
+喂	131
+晃	131
+烨	131
+瞎	131
+?	130
+②	130
+ù	130
+扁	130
+斥	130
+斩	130
+媚	130
+糊	130
+坪	130
+遮	130
+瘾	130
+拾	130
+爹	130
+?	129
+汛	129
+磅	129
+扳	129
+侬	129
+牺	129
+醇	129
+砸	129
+缅	129
+苍	129
+菊	129
+?	128
+茵	128
+惧	128
+割	128
+麟	128
+盼	128
+?	127
+?	127
+?	127
+?	127
+仗	127
+溃	127
+瑰	127
+浆	127
+哑	127
+霉	127
+嘿	127
+?	126
+?	126
+桩	126
+蹈	126
+潭	126
+捉	126
+洒	126
+饿	126
+玫	126
+哒	126
+滞	125
+枝	125
+催	125
+雯	125
+?	124
+?	124
+坠	124
+痕	124
+渗	123
+舱	123
+筱	123
+刹	123
+沦	123
+萧	123
+趁	123
+埋	123
+?	122
+厕	122
+脖	122
+哟	122
+?	121
+Ρ	121
+?	121
+?	121
+栅	121
+颈	121
+宙	121
+醋	121
+蝶	121
+踝	121
+?	120
+?	120
+?	120
+绽	120
+佬	120
+脆	120
+埔	120
+ζ	119
+?	119
+?	119
+?	119
+赁	119
+葩	119
+沫	119
+狭	119
+倩	119
+雌	119
+?	118
+?	118
+狄	118
+蒜	118
+湃	118
+犀	118
+冕	118
+蛇	118
+羡	118
+窟	118
+郅	117
+溯	117
+崛	117
+厘	117
+骄	117
+熬	117
+呱	117
+?	116
+?	116
+?	116
+嘲	116
+芬	116
+汹	116
+?	115
+?	115
+?	115
+髦	115
+捂	115
+罕	115
+芭	115
+绒	115
+淋	115
+捡	115
+臭	115
+芦	115
+?	114
+?	114
+?	114
+峻	114
+糟	114
+匙	114
+摊	114
+衬	114
+熙	114
+丘	114
+ι	113
+?	113
+雇	113
+茄	113
+啤	113
+闺	113
+?	112
+?	112
+浸	112
+飙	112
+腺	112
+呐	112
+拌	112
+搅	112
+?	111
+?	111
+?	111
+?	111
+?	111
+揽	111
+镀	111
+愧	111
+咳	111
+璐	111
+楠	111
+?	110
+ɡ	110
+吾	110
+杉	110
+锤	110
+挫	110
+巷	110
+?	109
+?	108
+诠	108
+朴	108
+腊	108
+禹	108
+灌	108
+?	107
+?	107
+磬	107
+铲	107
+辱	107
+吞	107
+葬	107
+杏	107
+悟	107
+1	106
+瞻	106
+?	106
+玥	106
+粥	106
+酮	106
+贱	106
+?	105
+?	105
+芃	105
+拯	105
+甸	105
+枢	105
+鞍	105
+柿	105
+澜	105
+殿	105
+闫	105
+砂	105
+誓	105
+陨	105
+?	104
+?	104
+e	104
+罢	104
+嗨	104
+饥	104
+?	103
+?	103
+?	103
+?	103
+?	103
+郊	103
+趟	103
+?	102
+?	102
+遏	102
+隧	102
+肇	102
+泌	102
+淫	102
+哗	102
+?	101
+?	101
+o	101
+?	101
+?	101
+慨	101
+燥	101
+缠	101
+粘	101
+躲	101
+疵	101
+?	100
+匪	100
+钩	100
+钥	100
+壤	100
+缸	100
+郫	99
+←	99
+?	99
+?	99
+?	99
+③	99
+歧	99
+郡	99
+粹	99
+旱	99
+焰	99
+崩	99
+臣	99
+哄	99
+娶	99
+茴	98
+?	98
+?	98
+а	98
+?	98
+籁	98
+姬	98
+彤	98
+寂	98
+?	97
+?	97
+?	97
+彦	97
+恭	97
+?	96
+?	96
+?	96
+?	96
+?	96
+滮	96
+垒	96
+邪	96
+扔	96
+?	95
+瞩	95
+?	95
+?	95
+侈	95
+巢	95
+勘	95
+?	94
+蕴	94
+菁	94
+筋	94
+沁	94
+彬	94
+翠	94
+?	93
+?	93
+?	93
+?	93
+?	93
+?	93
+酿	93
+壬	93
+宸	93
+咕	93
+绷	93
+龚	93
+戚	93
+俞	93
+菇	93
+兜	93
+靓	92
+沂	92
+谎	92
+迭	91
+凹	91
+瞄	91
+贬	91
+邵	91
+霖	91
+芸	91
+?	90
+￥	90
+祛	90
+蔓	90
+辟	90
+噪	90
+韵	90
+殴	90
+挨	90
+贼	90
+ò	89
+?	89
+俯	89
+焕	89
+壶	89
+昕	89
+栖	89
+钦	89
+躁	89
+恼	89
+蝴	89
+驴	89
+?	88
+?	88
+?	88
+?	88
+?	88
+膳	88
+蔚	88
+橙	88
+虐	88
+葱	88
+婉	88
+喧	88
+雍	88
+?	87
+?	87
+?	87
+恺	87
+撰	87
+瞒	87
+撮	87
+茫	87
+跪	87
+窦	87
+荆	87
+羞	87
+罐	87
+?	86
+④	86
+撩	86
+沸	86
+晗	86
+β	85
+ν	85
+?	85
+丨	85
+?	85
+麓	85
+悍	85
+牲	85
+漆	85
+奸	85
+?	84
+?	84
+?	84
+Я	84
+?	84
+／	84
+谌	84
+荧	84
+畏	84
+遂	84
+敞	84
+倪	84
+妍	84
+哀	84
+坝	84
+渣	84
+畜	84
+嫂	84
+?	83
+宛	83
+惕	83
+磷	83
+掐	83
+聋	83
+炜	83
+溶	83
+雀	83
+赌	83
+?	82
+?	82
+贿	82
+柄	82
+侃	82
+逝	82
+昊	82
+梨	82
+绚	82
+肆	82
+?	81
+?	81
+?	81
+?	81
+?	81
+滕	81
+睹	81
+竖	81
+闵	81
+笨	81
+?	80
+?	80
+?	80
+?	80
+仕	80
+皓	80
+尧	80
+蟹	80
+痴	80
+倦	80
+?	79
+?	79
+昧	79
+丫	79
+?	78
+?	78
+?	78
+哨	78
+弧	78
+亨	78
+盯	78
+棍	78
+舅	78
+?	77
+?	77
+夯	77
+谜	77
+湛	77
+捣	77
+枕	77
+邹	77
+屯	77
+虾	77
+钉	77
+葵	77
+蚊	77
+簧	77
+′	76
+?	76
+仯	76
+?	76
+篷	76
+扒	76
+辜	76
+枫	76
+佼	76
+?	75
+?	75
+帷	75
+?	75
+?	75
+犸	75
+韧	75
+沛	75
+勉	75
+阜	75
+歇	75
+绳	75
+氓	75
+讴	74
+?	74
+ü	74
+?	74
+苛	74
+?	74
+?	74
+皖	74
+渭	74
+黏	74
+遣	74
+硅	74
+婧	74
+刃	74
+甬	74
+邕	74
+隙	74
+寞	74
+晤	73
+炽	73
+?	73
+?	73
+?	73
+?	73
+鲸	73
+懈	73
+栽	73
+漠	73
+魄	73
+脊	73
+鲍	73
+澄	73
+椎	73
+麒	73
+娥	73
+芹	73
+焉	73
+唧	73
+邺	72
+?	72
+肢	72
+玖	72
+恤	72
+弦	72
+浑	72
+咸	72
+?	71
+?	71
+?	71
+?	71
+?	71
+?	71
+猿	71
+髓	71
+庚	71
+牡	71
+闸	71
+喻	71
+戳	71
+咒	71
+笋	71
+?	70
+棺	70
+氨	70
+垦	70
+蔽	70
+邢	70
+槐	70
+铬	69
+?	69
+?	69
+裔	69
+矶	69
+驭	69
+痰	69
+讳	69
+锣	69
+琅	69
+搁	69
+憋	69
+拢	69
+郝	69
+骆	69
+桐	69
+喉	69
+儒	69
+扇	69
+屠	69
+暇	69
+钾	69
+谅	69
+踊	69
+怂	69
+★	68
+?	68
+?	68
+?	68
+?	68
+阐	68
+俪	68
+毯	68
+匆	68
+扛	68
+黔	68
+茜	68
+妃	68
+?	67
+?	67
+?	67
+?	67
+?	67
+橱	67
+阀	67
+蹭	67
+淄	67
+坷	67
+?	66
+?	66
+?	66
+?	66
+?	66
+?	66
+舶	66
+衢	66
+悼	66
+姣	66
+臻	66
+铅	66
+驳	66
+叛	66
+庙	66
+捏	66
+?	65
+?	65
+豸	65
+掷	65
+顽	65
+喘	65
+甄	65
+锈	65
+?	64
+?	64
+?	64
+＋	64
+?	64
+?	64
+咽	64
+挪	64
+啸	64
+丛	64
+煎	64
+饲	64
+暧	64
+靖	64
+筒	64
+?	63
+?	63
+?	63
+?	63
+?	63
+怼	63
+°	63
+诵	63
+蝎	63
+鲨	63
+逮	63
+巫	63
+蕉	63
+襄	63
+罻	62
+⑤	62
+?	62
+漓	62
+苇	62
+囚	62
+栈	62
+毋	62
+帘	62
+肘	62
+饶	62
+?	61
+珑	61
+帜	61
+佑	61
+颂	61
+壹	61
+穴	61
+沾	61
+渲	61
+?	60
+赣	60
+?	60
+?	60
+钮	60
+榄	60
+缔	60
+摹	60
+羯	60
+梭	60
+奎	60
+耻	60
+汾	60
+卑	60
+吼	60
+醛	60
+趴	60
+恿	60
+湉	59
+?	59
+慷	59
+?	59
+?	59
+?	59
+?	59
+?	59
+?	59
+慑	59
+盎	59
+?	59
+骸	59
+酵	59
+稻	59
+膀	59
+矮	59
+喔	59
+膊	59
+?	58
+?	58
+?	58
+?	58
+烹	58
+?	58
+泻	58
+翟	58
+劈	58
+嘻	58
+葫	58
+悚	57
+橄	57
+孵	57
+酶	57
+叨	57
+枯	57
+缉	57
+漳	57
+宰	57
+雁	57
+塌	57
+玮	57
+琐	57
+霄	57
+嚣	57
+宵	57
+胳	57
+屌	57
+?	56
+?	56
+?	56
+?	56
+?	56
+厄	56
+犁	56
+殷	56
+讽	56
+睫	56
+橡	56
+嘎	56
+妞	56
+祠	56
+?	55
+?	55
+?	55
+α	55
+?	55
+遴	55
+呕	55
+涩	55
+冉	55
+愚	55
+捆	55
+廯	54
+?	54
+?	54
+梗	54
+娩	54
+殡	54
+砚	54
+蒿	54
+蚀	54
+煞	54
+矫	54
+疡	54
+丙	54
+膛	54
+烫	54
+岂	54
+裴	54
+喽	54
+?	53
+?	53
+跻	53
+芜	53
+喀	53
+肋	53
+邯	53
+陀	53
+栓	53
+挚	53
+耽	53
+榆	53
+蛛	53
+觅	53
+搂	53
+ī	52
+?	52
+?	52
+?	52
+?	52
+驿	52
+剿	52
+嗽	52
+凳	52
+墅	52
+棕	52
+���	52
+榴	52
+栗	52
+窘	52
+狡	52
+锚	51
+?	51
+2	51
+郸	51
+漾	51
+罄	51
+潍	51
+彝	51
+桓	51
+泵	51
+懿	51
+黛	51
+梓	51
+曰	51
+穹	51
+炯	51
+?	50
+⑥	50
+?	50
+?	50
+?	50
+玺	50
+灶	50
+垄	50
+琥	50
+刁	50
+泸	50
+煲	50
+蝉	50
+兮	50
+颅	50
+韬	50
+喇	50
+狸	50
+蛙	50
+翅	50
+匈	50
+茹	50
+褶	50
+?	49
+顷	49
+?	49
+徊	49
+拭	49
+筐	49
+嵌	49
+绊	49
+屿	49
+呛	49
+绅	49
+柠	49
+叮	49
+炖	49
+鹃	49
+屎	49
+猾	49
+揉	49
+跤	49
+?	48
+t	48
+徘	48
+?	48
+妊	48
+瀑	48
+衅	48
+棵	48
+嚼	48
+喵	48
+汕	48
+犬	48
+甫	48
+痘	48
+禧	48
+鲅	48
+?	47
+娠	47
+?	47
+?	47
+?	47
+?	47
+?	47
+?	47
+?	47
+?	47
+沥	47
+恳	47
+哺	47
+嗓	47
+冈	47
+樊	47
+禾	47
+皱	47
+蘑	47
+哼	47
+?	46
+廓	46
+?	46
+?	46
+?	46
+圭	46
+θ	46
+裬	46
+?	46
+鼾	46
+磋	46
+冶	46
+锯	46
+娅	46
+弈	46
+讶	46
+拱	46
+烘	46
+朽	46
+攒	46
+碌	46
+猥	46
+啪	46
+?	45
+?	45
+	45
+硒	45
+秤	45
+汐	45
+囧	45
+剔	45
+酋	45
+鸥	45
+蚕	45
+荔	45
+镶	45
+棠	45
+瞧	45
+祈	45
+皂	45
+痞	45
+?	44
+?	44
+■	44
+礁	44
+崎	44
+敛	44
+崭	44
+亟	44
+秆	44
+藕	44
+淼	44
+寥	44
+勺	44
+孜	44
+簇	44
+淹	44
+豚	44
+鸽	44
+斧	44
+伺	44
+慵	44
+龟	44
+?	43
+?	43
+薏	43
+?	43
+膨	43
+撬	43
+瘫	43
+嫣	43
+傍	43
+覃	43
+聆	43
+磕	43
+?	42
+?	42
+槿	42
+?	42
+?	42
+?	42
+秸	42
+猝	42
+诽	42
+谤	42
+淤	42
+飚	42
+摧	42
+篡	42
+窜	42
+匿	42
+隋	42
+檬	42
+撸	42
+焊	42
+钰	42
+俭	42
+浇	42
+甯	41
+颌	41
+阚	41
+惫	41
+?	41
+硢	41
+?	41
+?	41
+?	41
+?	41
+?	41
+捅	41
+?	41
+?	41
+忴	41
+?	41
+?	41
+?	41
+氢	41
+驯	41
+沅	41
+涿	41
+缀	41
+庐	41
+弓	41
+寅	41
+稽	41
+硫	41
+韶	41
+咧	41
+潢	41
+Ψ	40
+?	40
+?	40
+僚	40
+畴	40
+钛	40
+筝	40
+祁	40
+豁	40
+坟	40
+狙	40
+聂	40
+濮	40
+埠	40
+嘟	40
+毂	39
+诟	39
+?	39
+汲	39
+菅	39
+袱	39
+夭	39
+怠	39
+嘱	39
+阮	39
+糙	39
+晟	39
+炳	39
+叭	39
+驼	39
+棱	39
+筷	39
+袂	38
+?	38
+?	38
+?	38
+?	38
+碱	38
+?	38
+?	38
+?	38
+?	38
+?	38
+‰	38
+膺	38
+杞	38
+糗	38
+铸	38
+瓢	38
+窥	38
+缤	38
+鞭	38
+懵	38
+佣	38
+敷	38
+禅	38
+擒	38
+鳞	38
+嘞	38
+?	37
+黯	37
+?	37
+郴	37
+è	37
+ヽ	37
+汀	37
+?	37
+莺	37
+乍	37
+缭	37
+?	37
+羁	37
+竣	37
+夷	37
+蜕	37
+斐	37
+厮	37
+跆	37
+梧	37
+柚	37
+躯	37
+嗜	37
+奕	37
+稚	37
+琛	37
+杖	37
+奴	37
+η	36
+梵	36
+?	36
+?	36
+?	36
+?	36
+?	36
+?	36
+倔	36
+瀚	36
+诡	36
+焚	36
+娓	36
+佰	36
+乾	36
+抠	36
+恍	36
+拽	36
+盔	36
+暮	36
+瑾	36
+?	35
+飨	35
+?	35
+?	35
+?	35
+?	35
+辡	35
+↓	35
+?	35
+珏	35
+缪	35
+畸	35
+楷	35
+袍	35
+猩	35
+诛	35
+邸	35
+戛	35
+姗	35
+隶	35
+畔	35
+蜘	35
+姊	35
+犊	35
+唉	35
+鳕	35
+籭	34
+骐	34
+?	34
+?	34
+?	34
+?	34
+?	34
+咘	34
+媲	34
+讪	34
+梢	34
+靡	34
+靶	34
+饽	34
+嘘	34
+旷	34
+锌	34
+笛	34
+陡	34
+芪	34
+琢	34
+眸	34
+糯	34
+朔	34
+碾	34
+淇	34
+僻	34
+舔	34
+『	33
+』	33
+?	33
+觑	33
+?	33
+?	33
+?	33
+?	33
+?	33
+?	33
+?	33
+?	33
+濒	33
+?	33
+с	33
+蟒	33
+亵	33
+翩	33
+芮	33
+荟	33
+耘	33
+僧	33
+绮	33
+抒	33
+煜	33
+眩	33
+虞	33
+岐	33
+唠	33
+颐	33
+歹	33
+溺	33
+禄	33
+啃	33
+紊	33
+馍	33
+唔	33
+珈	32
+蠕	32
+?	32
+淶	32
+?	32
+?	32
+?	32
+?	32
+ˉ	32
+阪	32
+洼	32
+谴	32
+帧	32
+瞰	32
+恙	32
+臧	32
+氮	32
+妩	32
+鳅	32
+榨	32
+茧	32
+檀	32
+幌	32
+琊	32
+鸦	32
+娄	32
+巍	32
+惰	32
+彪	32
+榕	32
+瞅	32
+遐	32
+颊	32
+诶	32
+呃	32
+猷	31
+灼	31
+?	31
+é	31
+?	31
+?	31
+?	31
+Ь	31
+□	31
+裘	31
+猕	31
+淳	31
+辫	31
+徕	31
+掠	31
+郜	31
+株	31
+钳	31
+绯	31
+窖	31
+胚	31
+靳	31
+冥	31
+亢	31
+搓	31
+蹦	31
+撅	31
+?	30
+腥	30
+抉	30
+?	30
+?	30
+|	30
+?	30
+?	30
+嵥	30
+?	30
+?	30
+▼	30
+溉	30
+褚	30
+倘	30
+俨	30
+屹	30
+碟	30
+峥	30
+絮	30
+汶	30
+翡	30
+贞	30
+浊	30
+吟	30
+呦	30
+缇	29
+辄	29
+皙	29
+鏖	29
+?	29
+?	29
+?	29
+?	29
+?	29
+塣	29
+顢	29
+?	29
+锂	29
+赃	29
+枸	29
+烯	29
+硝	29
+曜	29
+堤	29
+仆	29
+熄	29
+滁	29
+倚	29
+揣	29
+祷	29
+鳝	29
+赐	29
+陋	29
+侄	29
+陂	29
+崑	28
+羺	28
+?	28
+?	28
+棘	28
+?	28
+?	28
+塢	28
+萃	28
+?	28
+?	28
+荼	28
+?	28
+?	28
+酣	28
+惋	28
+昱	28
+睁	28
+蔗	28
+峙	28
+伶	28
+蚌	28
+脐	28
+藤	28
+蒲	28
+趾	28
+痒	28
+疤	28
+钠	27
+孚	27
+?	27
+?	27
+?	27
+跷	27
+毗	27
+?	27
+鄞	27
+噱	27
+炙	27
+噩	27
+栾	27
+沽	27
+椰	27
+瓣	27
+鳄	27
+蝙	27
+蝠	27
+惶	27
+岔	27
+乞	27
+蹄	27
+荫	27
+跛	27
+?	26
+☆	26
+±	26
+?	26
+?	26
+?	26
+烊	26
+邃	26
+侥	26
+攘	26
+?	26
+轧	26
+桢	26
+矢	26
+萎	26
+魁	26
+洵	26
+蜗	26
+蜡	26
+璨	26
+姝	26
+穗	26
+渺	26
+惟	26
+莆	26
+泣	26
+鄙	26
+磺	26
+暄	26
+Τ	25
+?	25
+?	25
+?	25
+?	25
+?	25
+?	25
+擂	25
+?	25
+斛	25
+?	25
+涞	25
+汝	25
+辗	25
+殆	25
+砌	25
+邑	25
+坂	25
+峪	25
+剁	25
+窑	25
+岚	25
+芋	25
+鹞	24
+慜	24
+?	24
+镁	24
+ī	24
+?	24
+Ａ	24
+?	24
+?	24
+睦	24
+滇	24
+Ｂ	24
+饪	24
+涝	24
+辙	24
+羹	24
+莘	24
+匮	24
+拂	24
+幢	24
+熠	24
+胺	24
+璀	24
+缕	24
+辍	24
+肴	24
+澈	24
+��	24
+嚷	24
+洱	24
+咎	24
+褒	24
+襟	24
+涓	24
+伢	24
+?	23
+愫	23
+麾	23
+佟	23
+?	23
+?	23
+ì	23
+鹈	23
+鹕	23
+砒	23
+胰	23
+?	23
+ξ	23
+呷	23
+弛	23
+俘	23
+颓	23
+娆	23
+耸	23
+嫉	23
+侮	23
+璇	23
+牟	23
+尉	23
+驹	23
+濑	23
+?	22
+匣	22
+珲	22
+?	22
+鱦	22
+Ц	22
+?	22
+昙	22
+晾	22
+沮	22
+瞪	22
+窍	22
+藻	22
+沐	22
+秧	22
+酝	22
+飒	22
+窒	22
+羔	22
+蓓	22
+楞	22
+馒	22
+酥	22
+芙	22
+屑	22
+啧	22
+猖	22
+吮	22
+揍	22
+吖	22
+楂	21
+宕	21
+?	21
+?	21
+?	21
+Ⅴ	21
+?	21
+缜	21
+Ｐ	21
+悖	21
+罹	21
+稠	21
+柑	21
+Ｃ	21
+飩	21
+?	21
+?	21
+腱	21
+矣	21
+酉	21
+樟	21
+釉	21
+抨	21
+韭	21
+苟	21
+堕	21
+疸	21
+烁	21
+桔	21
+烽	21
+滔	21
+庵	21
+蝇	21
+蕊	21
+枭	21
+?	20
+?	20
+钜	20
+?	20
+?	20
+?	20
+囤	20
+?	20
+?	20
+?	20
+?	20
+?	20
+?	20
+｀	20
+阱	20
+?	20
+迸	20
+?	20
+?	20
+瑛	20
+―	20
+?	20
+?	20
+蹿	20
+迥	20
+苓	20
+籽	20
+骋	20
+崂	20
+哮	20
+诫	20
+挠	20
+粪	20
+湄	20
+嬉	20
+卉	20
+煽	20
+晦	20
+铂	20
+漱	20
+妒	20
+掺	20
+揪	20
+龈	20
+铌	19
+戟	19
+邬	19
+硣	19
+?	19
+蹊	19
+亳	19
+譬	19
+斡	19
+淖	19
+匝	19
+?	19
+峭	19
+χ	19
+痪	19
+赎	19
+喆	19
+搀	19
+酌	19
+貂	19
+癖	19
+嵩	19
+茬	19
+礴	19
+肽	19
+寰	19
+憨	19
+隽	19
+昼	19
+铉	19
+拎	19
+鲤	19
+瞿	19
+饺	19
+峨	19
+剐	19
+荚	19
+鳖	19
+?	18
+岖	18
+漕	18
+珐	18
+洢	18
+贲	18
+?	18
+弋	18
+嗅	18
+?	18
+?	18
+?	18
+蜥	18
+腑	18
+腭	18
+─	18
+歆	18
+骼	18
+?	18
+?	18
+颤	18
+赘	18
+镰	18
+旖	18
+徙	18
+煦	18
+瑚	18
+丞	18
+惬	18
+饷	18
+嗦	18
+晔	18
+婿	18
+秃	18
+懦	18
+逍	18
+竿	18
+啰	18
+珂	18
+癫	18
+墩	18
+浣	18
+掰	18
+醺	18
+姥	18
+臃	18
+嬷	18
+崽	18
+妓	18
+噶	18
+?	17
+?	17
+?	17
+氪	17
+＆	17
+?	17
+涪	17
+?	17
+?	17
+?	17
+?	17
+幤	17
+	17
+?	17
+?	17
+凿	17
+鰸	17
+搡	17
+栩	17
+≠	17
+塱	17
+戮	17
+灸	17
+扼	17
+泠	17
+藉	17
+痊	17
+弩	17
+簿	17
+刨	17
+芥	17
+镌	17
+槟	17
+碘	17
+铨	17
+葆	17
+侍	17
+昀	17
+泾	17
+鱿	17
+脯	17
+惚	17
+粑	17
+蛀	17
+嗲	17
+叽	17
+嘚	17
+朕	17
+鸢	17
+骢	16
+蛳	16
+洙	16
+?	16
+鹌	16
+鹑	16
+唾	16
+苋	16
+?	16
+岌	16
+噬	16
+?	16
+?	16
+偃	16
+鷢	16
+?	16
+谧	16
+瞠	16
+?	16
+祉	16
+跋	16
+颍	16
+?	16
+舸	16
+惭	16
+怯	16
+酚	16
+酯	16
+綦	16
+涤	16
+挟	16
+漪	16
+烙	16
+纫	16
+毓	16
+腌	16
+椿	16
+盹	16
+咐	16
+皑	16
+咙	16
+氟	16
+拷	16
+缙	16
+咩	16
+啵	16
+￡	15
+?	15
+伉	15
+釜	15
+昇	15
+鞣	15
+垩	15
+?	15
+?	15
+珮	15
+崆	15
+峒	15
+佚	15
+笃	15
+蜈	15
+蚣	15
+摒	15
+?	15
+徭	15
+鴦	15
+哽	15
+铐	15
+廿	15
+?	15
+◆	15
+魇	15
+缮	15
+潺	15
+钝	15
+鲢	15
+緢	15
+淌	15
+恬	15
+皋	15
+镍	15
+缚	15
+踵	15
+霁	15
+氯	15
+憬	15
+轶	15
+嘹	15
+雏	15
+卤	15
+饵	15
+殒	15
+孰	15
+圩	15
+镂	15
+涟	15
+褐	15
+妤	15
+隘	15
+帼	15
+莴	15
+箍	15
+粱	15
+贰	15
+哩	15
+诅	15
+懋	14
+萤	14
+嬅	14
+玏	14
+赟	14
+硶	14
+咄	14
+拗	14
+a	14
+冗	14
+?	14
+壩	14
+嫦	14
+?	14
+邗	14
+?	14
+讹	14
+?	14
+?	14
+缨	14
+?	14
+?	14
+齨	14
+埇	14
+啼	14
+◎	14
+咀	14
+旎	14
+〉	14
+涣	14
+墒	14
+蜿	14
+蜒	14
+?	14
+?	14
+劾	14
+哆	14
+馥	14
+酪	14
+翊	14
+憧	14
+鄱	14
+诬	14
+瓯	14
+骞	14
+岱	14
+禀	14
+羌	14
+俐	14
+苔	14
+柬	14
+绞	14
+锥	14
+嚎	14
+庇	14
+霏	14
+熹	14
+壕	14
+悸	14
+槌	14
+诩	14
+遛	14
+搪	14
+殃	14
+疮	14
+眯	14
+菩	14
+舜	14
+豌	14
+婶	14
+浔	14
+喃	14
+糍	14
+蛹	14
+溴	13
+於	13
+曳	13
+?	13
+芡	13
+瑀	13
+?	13
+?	13
+邂	13
+逅	13
+?	13
+砥	13
+褪	13
+淅	13
+淆	13
+梆	13
+逵	13
+?	13
+?	13
+?	13
+?	13
+鶽	13
+?	13
+?	13
+桱	13
+婺	13
+?	13
+?	13
+莠	13
+?	13
+砺	13
+鵽	13
+亥	13
+饕	13
+餮	13
+卯	13
+瞭	13
+〈	13
+?	13
+蛰	13
+奂	13
+?	13
+鳤	13
+蹤	13
+拣	13
+札	13
+妪	13
+馁	13
+桨	13
+舵	13
+莽	13
+趸	13
+瞳	13
+诲	13
+幡	13
+哉	13
+戎	13
+焱	13
+拧	13
+桦	13
+昵	13
+晏	13
+撇	13
+炬	13
+衩	13
+彷	13
+捶	13
+叼	13
+痹	13
+辊	13
+婵	13
+恽	12
+孃	12
+?	12
+∕	12
+?	12
+戌	12
+芈	12
+桅	12
+糡	12
+?	12
+茯	12
+?	12
+?	12
+?	12
+?	12
+侏	12
+蕨	12
+斓	12
+ū	12
+﹟	12
+?	12
+滭	12
+罡	12
+锭	12
+Ъ	12
+?	12
+?	12
+?	12
+泿	12
+Ф	12
+?	12
+€	12
+伎	12
+?	12
+谩	12
+沼	12
+佹	12
+﹑	12
+腼	12
+腆	12
+锷	12
+泞	12
+忱	12
+诧	12
+檐	12
+盏	12
+榻	12
+璞	12
+嘀	12
+泗	12
+忐	12
+忑	12
+烛	12
+浜	12
+咚	12
+霓	12
+疹	12
+榈	12
+菏	12
+沭	12
+雹	12
+坯	12
+咤	12
+拄	12
+璋	12
+咔	12
+瞌	12
+笠	12
+矜	12
+睇	12
+凋	12
+煊	12
+妾	12
+?	11
+桀	11
+?	11
+籾	11
+?	11
+狰	11
+狞	11
+嚏	11
+嗖	11
+茌	11
+?	11
+?	11
+鹦	11
+榷	11
+谚	11
+?	11
+缥	11
+缈	11
+枷	11
+谛	11
+?	11
+?	11
+¤	11
+?	11
+烩	11
+璟	11
+漰	11
+?	11
+?	11
+?	11
+?	11
+?	11
+耄	11
+耋	11
+?	11
+堃	11
+?	11
+嘈	11
+復	11
+?	11
+垢	11
+茎	11
+沱	11
+?	11
+?	11
+眶	11
+?	11
+Ⅲ	11
+粳	11
+珉	11
+惮	11
+榛	11
+窈	11
+窕	11
+逞	11
+鲫	11
+兀	11
+阂	11
+卒	11
+犷	11
+恣	11
+虔	11
+嘶	11
+镖	11
+钵	11
+岷	11
+郧	11
+馅	11
+碉	11
+祀	11
+疚	11
+腮	11
+茸	11
+涕	11
+镐	11
+苯	11
+蛤	11
+邛	11
+琨	11
+呜	11
+丐	11
+仨	11
+荻	10
+?	10
+?	10
+＊	10
+藜	10
+甙	10
+胱	10
+耦	10
+鹉	10
+匕	10
+Ｉ	10
+洦	10
+?	10
+浥	10
+拙	10
+挎	10
+?	10
+掖	10
+?	10
+矾	10
+簸	10
+酗	10
+莒	10
+?	10
+?	10
+?	10
+?	10
+?	10
+吋	10
+闱	10
+?	10
+綯	10
+?	10
+箔	10
+?	10
+渚	10
+?	10
+Т	10
+篓	10
+荤	10
+?	10
+眺	10
+猞	10
+猁	10
+?	10
+?	10
+氙	10
+漩	10
+婊	10
+寐	10
+丕	10
+凇	10
+稼	10
+铮	10
+纶	10
+泓	10
+粟	10
+潞	10
+鹭	10
+竺	10
+婪	10
+荞	10
+裳	10
+惦	10
+椭	10
+髋	10
+吱	10
+拈	10
+愣	10
+琉	10
+沌	10
+蹶	10
+蔺	10
+潼	10
+徨	10
+嗑	10
+甥	10
+芷	10
+狒	10
+恁	10
+俍	9
+舷	9
+恫	9
+?	9
+▌	9
+?	9
+镕	9
+?	9
+亞	9
+梼	9
+杌	9
+踞	9
+赅	9
+⑦	9
+绫	9
+诙	9
+缎	9
+?	9
+?	9
+遨	9
+狩	9
+Ｔ	9
+唏	9
+?	9
+舫	9
+?	9
+笈	9
+螈	9
+黠	9
+罨	9
+繫	9
+熔	9
+?	9
+鼮	9
+?	9
+?	9
+?	9
+淢	9
+?	9
+?	9
+吒	9
+鲵	9
+篑	9
+樾	9
+?	9
+榭	9
+齫	9
+桰	9
+?	9
+?	9
+睢	9
+鞘	9
+Ｄ	9
+吝	9
+谙	9
+摃	9
+渍	9
+?	9
+仞	9
+琶	9
+叱	9
+溏	9
+萦	9
+赂	9
+斟	9
+渎	9
+抿	9
+恕	9
+隅	9
+濡	9
+拇	9
+扉	9
+屉	9
+圃	9
+锵	9
+袄	9
+厥	9
+劵	9
+迩	9
+馋	9
+祯	9
+粽	9
+呲	9
+镯	9
+幺	9
+隍	9
+捎	9
+孽	9
+噻	9
+?	8
+濨	8
+?	8
+伱	8
+?	8
+?	8
+場	8
+忪	8
+逯	8
+?	8
+嬛	8
+阖	8
+锏	8
+叒	8
+ん	8
+鞅	8
+铩	8
+窨	8
+＾	8
+?	8
+笕	8
+谬	8
+滺	8
+鍊	8
+栎	8
+芩	8
+桎	8
+梏	8
+?	8
+臼	8
+?	8
+?	8
+蟾	8
+龋	8
+罔	8
+滹	8
+?	8
+蝾	8
+糅	8
+?	8
+?	8
+飺	8
+獗	8
+鞠	8
+?	8
+羰	8
+涼	8
+綼	8
+鱨	8
+?	8
+?	8
+嶂	8
+?	8
+鍦	8
+棰	8
+揄	8
+?	8
+鹊	8
+粼	8
+?	8
+у	8
+Ⅳ	8
+岙	8
+臆	8
+桷	8
+炊	8
+辘	8
+吏	8
+羚	8
+鳍	8
+泱	8
+娣	8
+熨	8
+垚	8
+溅	8
+褥	8
+噜	8
+铰	8
+撂	8
+凛	8
+婕	8
+绾	8
+翌	8
+焙	8
+丶	8
+螃	8
+燎	8
+忒	8
+溧	8
+篆	8
+噗	8
+踹	8
+妳	8
+浒	8
+锰	8
+陛	8
+嫖	8
+璜	8
+绢	8
+捻	8
+瘸	8
+砀	7
+Ｓ	7
+铢	7
+將	7
+缄	7
+胤	7
+旻	7
+祎	7
+脍	7
+谲	7
+洌	7
+槃	7
+№	7
+?	7
+⑧	7
+蹴	7
+?	7
+≥	7
+傀	7
+儡	7
+?	7
+＂	7
+恪	7
+泩	7
+杲	7
+?	7
+?	7
+覴	7
+颧	7
+磐	7
+瘠	7
+?	7
+?	7
+?	7
+锲	7
+楹	7
+蛾	7
+?	7
+潫	7
+壺	7
+﹒	7
+仚	7
+罂	7
+?	7
+?	7
+?	7
+濪	7
+黹	7
+瀛	7
+浚	7
+?	7
+?	7
+?	7
+?	7
+?	7
+?	7
+籩	7
+?	7
+?	7
+-	7
+捺	7
+瑙	7
+鏃	7
+肮	7
+潨	7
+淦	7
+艏	7
+亘	7
+鲻	7
+矗	7
+鄯	7
+?	7
+籣	7
+馏	7
+赦	7
+琵	7
+胫	7
+膑	7
+孪	7
+窿	7
+嗒	7
+痧	7
+枇	7
+杷	7
+咫	7
+吭	7
+郯	7
+铆	7
+睬	7
+烬	7
+灞	7
+殇	7
+叁	7
+迦	7
+硼	7
+袒	7
+珞	7
+抡	7
+胭	7
+茁	7
+猬	7
+濠	7
+裆	7
+螂	7
+崃	7
+涧	7
+痔	7
+奚	7
+痫	7
+笙	7
+疙	7
+瘩	7
+玷	7
+胛	7
+潦	7
+坨	7
+卅	7
+钊	7
+蔑	7
+芍	7
+腋	7
+泔	6
+阕	6
+俾	6
+讣	6
+塷	6
+擘	6
+骛	6
+?	6
+劢	6
+鲹	6
+鷿	6
+?	6
+忡	6
+旌	6
+?	6
+?	6
+痼	6
+?	6
+揶	6
+?	6
+?	6
+達	6
+奪	6
+奧	6
+	6
+眈	6
+仵	6
+洇	6
+蹒	6
+跚	6
+遑	6
+嗤	6
+?	6
+?	6
+鹜	6
+?	6
+脓	6
+?	6
+皎	6
+?	6
+苣	6
+痨	6
+瘀	6
+黳	6
+?	6
+薰	6
+?	6
+?	6
+?	6
+碁	6
+蜍	6
+麂	6
+?	6
+涾	6
+?	6
+叩	6
+?	6
+?	6
+?	6
+赡	6
+讧	6
+跽	6
+?	6
+?	6
+?	6
+裨	6
+?	6
+?	6
+鹫	6
+掣	6
+鄢	6
+?	6
+?	6
+?	6
+?	6
+?	6
+硱	6
+?	6
+?	6
+?	6
+?	6
+霹	6
+雳	6
+?	6
+齭	6
+刍	6
+荛	6
+?	6
+畈	6
+?	6
+趨	6
+?	6
+Ｇ	6
+侗	6
+叕	6
+銆	6
+棤	6
+數	6
+纾	6
+匾	6
+?	6
+徇	6
+恃	6
+戰	6
+伫	6
+袅	6
+?	6
+?	6
+?	6
+讥	6
+喱	6
+哧	6
+邋	6
+遢	6
+肱	6
+拮	6
+澧	6
+瘪	6
+茱	6
+蕃	6
+钗	6
+惺	6
+怅	6
+攥	6
+捋	6
+黝	6
+痢	6
+禺	6
+飓	6
+匡	6
+癜	6
+汞	6
+踌	6
+躇	6
+崴	6
+嗝	6
+衙	6
+蔼	6
+淞	6
+悯	6
+岑	6
+诏	6
+矬	6
+艮	5
+漉	5
+?	5
+?	5
+?	5
+Ｍ	5
+↑	5
+蛊	5
+?	5
+	5
+冢	5
+?	5
+砵	5
+?	5
+隼	5
+?	5
+漣	5
+後	5
+聯	5
+爭	5
+勝	5
+琚	5
+澍	5
+佤	5
+檄	5
+匍	5
+匐	5
+躬	5
+嬴	5
+雒	5
+?	5
+?	5
+?	5
+孑	5
+喑	5
+禚	5
+?	5
+裺	5
+缰	5
+腩	5
+槲	5
+弰	5
+?	5
+?	5
+糿	5
+緹	5
+?	5
+?	5
+鷨	5
+?	5
+徫	5
+?	5
+廹	5
+壑	5
+烷	5
+?	5
+?	5
+?	5
+鷳	5
+顱	5
+?	5
+週	5
+?	5
+?	5
+赲	5
+?	5
+忏	5
+蛎	5
+弶	5
+怆	5
+?	5
+?	5
+簲	5
+?	5
+藩	5
+侭	5
+褂	5
+?	5
+?	5
+?	5
+浱	5
+?	5
+磩	5
+?	5
+?	5
+Ｏ	5
+?	5
+巽	5
+鷽	5
+鼰	5
+繤	5
+眨	5
+?	5
+垛	5
+聿	5
+垟	5
+?	5
+?	5
+?	5
+?	5
+佘	5
+蜴	5
+藐	5
+绌	5
+绛	5
+紝	5
+犵	5
+氳	5
+锛	5
+?	5
+?	5
+渥	5
+夙	5
+扞	5
+镚	5
+堑	5
+儆	5
+△	5
+诿	5
+?	5
+Ⅱ	5
+酰	5
+辋	5
+瓒	5
+篱	5
+蕙	5
+瞥	5
+钏	5
+铎	5
+圻	5
+摁	5
+埭	5
+拴	5
+珩	5
+硐	5
+铿	5
+芊	5
+墟	5
+蟑	5
+蘸	5
+榔	5
+黍	5
+耙	5
+茗	5
+盱	5
+眙	5
+翱	5
+扈	5
+瘟	5
+橇	5
+娼	5
+宥	5
+闰	5
+嗷	5
+吩	5
+稣	5
+勐	5
+浃	5
+肛	5
+侉	5
+荨	5
+啷	5
+龅	5
+呸	5
+幩	4
+谆	4
+Ⅵ	4
+锾	4
+?	4
+悢	4
+鲉	4
+Ｌ	4
+鹱	4
+兢	4
+骰	4
+敕	4
+羳	4
+蛟	4
+?	4
+?	4
+?	4
+?	4
+鰴	4
+?	4
+?	4
+齺	4
+?	4
+蹩	4
+綾	4
+?	4
+雲	4
+?	4
+薅	4
+純	4
+?	4
+é	4
+?	4
+對	4
+維	4
+無	4
+爾	4
+碴	4
+當	4
+壡	4
+樯	4
+?	4
+踽	4
+?	4
+觥	4
+Ｋ	4
+撷	4
+镝	4
+ㄞ	4
+檎	4
+こ	4
+ば	4
+は	4
+淬	4
+帚	4
+诨	4
+囿	4
+荏	4
+苒	4
+紴	4
+┃	4
+塭	4
+巿	4
+?	4
+?	4
+?	4
+?	4
+ｘ	4
+沓	4
+苷	4
+糱	4
+?	4
+?	4
+啬	4
+訇	4
+?	4
+弡	4
+痠	4
+懊	4
+溬	4
+?	4
+焯	4
+蜚	4
+轼	4
+?	4
+?	4
+攫	4
+?	4
+塶	4
+?	4
+峵	4
+砰	4
+汸	4
+徳	4
+岢	4
+–	4
+粣	4
+賿	4
+妁	4
+?	4
+?	4
+埌	4
+嘢	4
+昶	4
+緽	4
+?	4
+嗣	4
+麋	4
+婀	4
+繦	4
+碚	4
+鲼	4
+?	4
+汣	4
+粿	4
+?	4
+?	4
+踲	4
+邡	4
+沣	4
+陟	4
+?	4
+?	4
+?	4
+?	4
+塰	4
+飻	4
+?	4
+?	4
+?	4
+?	4
+跢	4
+翀	4
+憔	4
+悴	4
+峯	4
+篦	4
+?	4
+輸	4
+?	4
+漴	4
+?	4
+?	4
+岻	4
+殉	4
+贮	4
+峦	4
+稔	4
+鐢	4
+	4
+瀹	4
+氬	4
+	4
+鍙	4
+柟	4
+ゆ	4
+戠	4
+巼	4
+費	4
+湫	4
+坍	4
+蘖	4
+颡	4
+谑	4
+琍	4
+鸠	4
+愕	4
+啾	4
+杳	4
+耆	4
+栉	4
+?	4
+?	4
+忻	4
+粕	4
+骇	4
+弑	4
+掇	4
+鬓	4
+莅	4
+厝	4
+瑄	4
+虏	4
+弼	4
+蔷	4
+阙	4
+钨	4
+塬	4
+萸	4
+樵	4
+娉	4
+荃	4
+孢	4
+纭	4
+炀	4
+犟	4
+耷	4
+甭	4
+遁	4
+蹬	4
+潸	4
+铣	4
+鸾	4
+娲	4
+卞	4
+偕	4
+苞	4
+茉	4
+胥	4
+衲	4
+蠛	4
+荀	4
+墉	4
+痿	4
+铧	4
+淨	3
+?	3
+?	3
+饬	3
+嫡	3
+溱	3
+蚵	3
+铀	3
+钽	3
+佃	3
+孺	3
+Ｗ	3
+绥	3
+蟠	3
+?	3
+?	3
+?	3
+﹝	3
+﹞	3
+?	3
+?	3
+诣	3
+?	3
+?	3
+?	3
+帣	3
+?	3
+癿	3
+瓴	3
+內	3
+浽	3
+彨	3
+呎	3
+淝	3
+顰	3
+郦	3
+?	3
+徹	3
+榫	3
+麣	3
+滻	3
+铄	3
+儋	3
+挹	3
+曩	3
+孱	3
+瑪	3
+?	3
+?	3
+販	3
+?	3
+?	3
+?	3
+?	3
+?	3
+?	3
+?	3
+跬	3
+埈	3
+?	3
+愍	3
+刿	3
+Ｎ	3
+哌	3
+○	3
+?	3
+?	3
+?	3
+щ	3
+?	3
+徰	3
+裻	3
+轭	3
+?	3
+賬	3
+?	3
+會	3
+輪	3
+開	3
+勢	3
+撺	3
+暹	3
+谶	3
+掴	3
+黰	3
+侳	3
+?	3
+荜	3
+踯	3
+躅	3
+憎	3
+晞	3
+?	3
+?	3
+霰	3
+辿	3
+塚	3
+?	3
+幫	3
+?	3
+喙	3
+?	3
+︰	3
+の	3
+糜	3
+?	3
+?	3
+?	3
+?	3
+籡	3
+?	3
+?	3
+蔻	3
+▎	3
+祐	3
+粻	3
+?	3
+?	3
+?	3
+?	3
+?	3
+廤	3
+膈	3
+髲	3
+瘁	3
+辂	3
+時	3
+樨	3
+诜	3
+麸	3
+?	3
+?	3
+嘌	3
+呤	3
+?	3
+?	3
+?	3
+?	3
+?	3
+轡	3
+?	3
+膘	3
+擞	3
+钼	3
+荠	3
+蔫	3
+?	3
+濛	3
+罫	3
+佱	3
+?	3
+繺	3
+?	3
+?	3
+?	3
+?	3
+?	3
+鱾	3
+跶	3
+澫	3
+徵	3
+?	3
+?	3
+乺	3
+?	3
+?	3
+?	3
+?	3
+陲	3
+仄	3
+溥	3
+?	3
+?	3
+?	3
+舛	3
+檩	3
+?	3
+?	3
+?	3
+掬	3
+嶯	3
+?	3
+嵘	3
+?	3
+?	3
+?	3
+髯	3
+?	3
+铤	3
+?	3
+嵽	3
+?	3
+?	3
+峤	3
+纂	3
+鶼	3
+?	3
+棨	3
+辭	3
+櫻	3
+簪	3
+桉	3
+诤	3
+惘	3
+攸	3
+棩	3
+?	3
+衱	3
+?	3
+?	3
+?	3
+潻	3
+?	3
+?	3
+仧	3
+?	3
+缯	3
+绹	3
+?	3
+?	3
+嫘	3
+?	3
+?	3
+?	3
+?	3
+辶	3
+?	3
+?	3
+乫	3
+?	3
+?	3
+?	3
+侲	3
+弢	3
+黟	3
+阑	3
+瘘	3
+勍	3
+鎷	3
+灏	3
+併	3
+細	3
+鎮	3
+鏈	3
+鍔	3
+鐜	3
+浼	3
+皥	3
+	3
+鐞	3
+嗗	3
+	3
+	3
+绠	3
+＄	3
+悊	3
+呭	3
+鍜	3
+鍥	3
+屽	3
+鍗	3
+璁	3
+綉	3
+?	3
+?	3
+晢	3
+迂	3
+箩	3
+埂	3
+懑	3
+镭	3
+氡	3
+徜	3
+徉	3
+涸	3
+?	3
+裡	3
+憩	3
+袤	3
+戍	3
+?	3
+鴫	3
+牤	3
+龊	3
+锹	3
+秽	3
+噘	3
+谄	3
+鲛	3
+骧	3
+魍	3
+魉	3
+诳	3
+俸	3
+噎	3
+扥	3
+帛	3
+蚝	3
+鳗	3
+獠	3
+颉	3
+祟	3
+澹	3
+鲈	3
+呻	3
+獒	3
+漯	3
+骊	3
+囱	3
+砾	3
+掂	3
+蹂	3
+躏	3
+鹄	3
+轲	3
+瞑	3
+吆	3
+挞	3
+唬	3
+潆	3
+筠	3
+蹋	3
+婢	3
+腚	3
+鲶	3
+喏	3
+馄	3
+饨	3
+佢	3
+嗟	3
+尥	3
+仟	3
+埗	2
+?	2
+?	2
+硰	2
+訾	2
+?	2
+漶	2
+龌	2
+Ｖ	2
+鸸	2
+鹋	2
+俋	2
+谳	2
+冑	2
+馊	2
+饯	2
+晌	2
+橹	2
+谘	2
+峋	2
+?	2
+ê	2
+祗	2
+坜	2
+浬	2
+倭	2
+鲟	2
+畿	2
+驛	2
+诋	2
+辯	2
+≈	2
+癀	2
+?	2
+芎	2
+?	2
+號	2
+鰵	2
+?	2
+滟	2
+汻	2
+?	2
+疴	2
+瘴	2
+Ｆ	2
+?	2
+岣	2
+＃	2
+滷	2
+黩	2
+賤	2
+?	2
+軭	2
+媸	2
+綴	2
+蜻	2
+蜓	2
+?	2
+觊	2
+觎	2
+锳	2
+?	2
+?	2
+衭	2
+?	2
+廧	2
+?	2
+?	2
+妫	2
+?	2
+?	2
+?	2
+?	2
+?	2
+?	2
+?	2
+?	2
+?	2
+?	2
+?	2
+?	2
+?	2
+?	2
+趯	2
+?	2
+?	2
+?	2
+¨	2
+?	2
+挝	2
+莼	2
+唳	2
+?	2
+?	2
+?	2
+?	2
+銭	2
+踉	2
+跄	2
+?	2
+?	2
+涎	2
+蟬	2
+橫	2
+蘭	2
+風	2
+頭	2
+紅	2
+脫	2
+?	2
+?	2
+?	2
+?	2
+?	2
+?	2
+?	2
+?	2
+跖	2
+玓	2
+?	2
+纰	2
+邳	2
+?	2
+缶	2
+?	2
+衵	2
+紐	2
+倫	2
+軍	2
+過	2
+來	2
+錫	2
+魯	2
+麽	2
+侩	2
+馃	2
+齁	2
+咻	2
+畑	2
+湊	2
+?	2
+?	2
+?	2
+為	2
+掮	2
+蒹	2
+葭	2
+髻	2
+?	2
+愠	2
+牍	2
+傈	2
+僳	2
+獭	2
+匏	2
+跺	2
+?	2
+?	2
+?	2
+娑	2
+诃	2
+恻	2
+ū	2
+笆	2
+魑	2
+鼬	2
+蜃	2
+戬	2
+纣	2
+?	2
+?	2
+颏	2
+迢	2
+骥	2
+轣	2
+?	2
+鵰	2
+?	2
+?	2
+?	2
+?	2
+?	2
+?	2
+?	2
+廬	2
+廼	2
+迺	2
+?	2
+?	2
+紺	2
+喋	2
+濯	2
+玑	2
+玟	2
+啫	2
+噚	2
+?	2
+?	2
+?	2
+?	2
+?	2
+?	2
+?	2
+?	2
+?	2
+?	2
+?	2
+?	2
+?	2
+?	2
+?	2
+?	2
+覸	2
+?	2
+谮	2
+?	2
+痉	2
+挛	2
+瘢	2
+?	2
+馕	2
+?	2
+?	2
+?	2
+讚	2
+緻	2
+冇	2
+蹙	2
+酩	2
+酊	2
+姒	2
+妲	2
+琰	2
+痣	2
+锢	2
+箕	2
+甾	2
+峿	2
+魻	2
+?	2
+?	2
+蕹	2
+給	2
+肼	2
+?	2
+鵤	2
+?	2
+煨	2
+?	2
+佺	2
+晧	2
+驽	2
+怿	2
+飡	2
+俴	2
+?	2
+?	2
+?	2
+?	2
+?	2
+淴	2
+▽	2
+?	2
+?	2
+?	2
+籦	2
+?	2
+轺	2
+?	2
+?	2
+乩	2
+壨	2
+?	2
+?	2
+扪	2
+緶	2
+洹	2
+?	2
+?	2
+泮	2
+?	2
+滉	2
+?	2
+?	2
+颚	2
+嵭	2
+俿	2
+?	2
+義	2
+鼱	2
+罘	2
+蚱	2
+?	2
+?	2
+蠖	2
+脲	2
+?	2
+幄	2
+坻	2
+〇	2
+縺	2
+縲	2
+唆	2
+擀	2
+佝	2
+偻	2
+晁	2
+兖	2
+惴	2
+娭	2
+毑	2
+朦	2
+胧	2
+湤	2
+翪	2
+漼	2
+忨	2
+劭	2
+滆	2
+嵊	2
+沢	2
+戊	2
+邝	2
+浠	2
+績	2
+閰	2
+ㄧ	2
+嶇	2
+棣	2
+栭	2
+侀	2
+閫	2
+唴	2
+笌	2
+欐	2
+笟	2
+崗	2
+冧	2
+笓	2
+寮	2
+傚	2
+鍖	2
+眬	2
+姟	2
+佷	2
+鍐	2
+湴	2
+ㄥ	2
+屾	2
+祦	2
+氫	2
+鎯	2
+喌	2
+晫	2
+	2
+绉	2
+姩	2
+杩	2
+嗙	2
+滃	2
+簡	2
+缁	2
+戞	2
+岀	2
+愬	2
+鏂	2
+叡	2
+嗕	2
+堜	2
+ㄦ	2
+鍚	2
+鎶	2
+	2
+浗	2
+姙	2
+曚	2
+綅	2
+沖	2
+埸	2
+潔	2
+糠	2
+啕	2
+枨	2
+牦	2
+汩	2
+訚	2
+鰲	2
+偎	2
+証	2
+鸱	2
+阡	2
+骡	2
+笺	2
+锄	2
+玢	2
+咦	2
+侑	2
+邰	2
+锴	2
+樽	2
+螳	2
+羿	2
+嵋	2
+蛆	2
+巳	2
+砷	2
+岫	2
+泯	2
+豇	2
+嗔	2
+壅	2
+摞	2
+阄	2
+逑	2
+镊	2
+檫	2
+赝	2
+蓟	2
+锟	2
+斋	2
+汨	2
+癣	2
+朐	2
+蓑	2
+叻	2
+疝	2
+桁	2
+赳	2
+谔	2
+垣	2
+葚	2
+搵	2
+貔	2
+貅	2
+衮	2
+哔	2
+氤	2
+氲	2
+蜛	2
+襦	2
+蝌	2
+蚪	2
+疟	2
+鬟	2
+鸬	2
+鹚	2
+誊	2
+艄	2
+吶	2
+蝈	2
+祢	2
+滢	2
+郏	1
+沆	1
+瀣	1
+岿	1
+苈	1
+礜	1
+庾	1
+跹	1
+孓	1
+蚩	1
+偲	1
+峧	1
+踻	1
+裩	1
+絽	1
+裱	1
+闳	1
+町	1
+塾	1
+讷	1
+俚	1
+倥	1
+偬	1
+纛	1
+锱	1
+蔣	1
+皈	1
+溟	1
+箓	1
+胪	1
+啻	1
+暌	1
+鸵	1
+釆	1
+嚓	1
+俎	1
+庖	1
+畚	1
+蒨	1
+洄	1
+箫	1
+韡	1
+镞	1
+僭	1
+忿	1
+铳	1
+铵	1
+鲑	1
+囡	1
+収	1
+皿	1
+彡	1
+稷	1
+埝	1
+貽	1
+郢	1
+鳡	1
+跣	1
+郓	1
+鄄	1
+箬	1
+堋	1
+堽	1
+遒	1
+钣	1
+佶	1
+墀	1
+請	1
+應	1
+這	1
+個	1
+動	1
+揮	1
+獲	1
+儘	1
+隊	1
+頹	1
+搶	1
+兇	1
+們	1
+陣	1
+絕	1
+實	1
+飛	1
+腳	1
+懼	1
+鑽	1
+馬	1
+體	1
+夠	1
+曬	1
+貝	1
+頓	1
+盧	1
+帶	1
+盤	1
+證	1
+怔	1
+渟	1
+貘	1
+荪	1
+郞	1
+賽	1
+敵	1
+頂	1
+機	1
+謂	1
+繼	1
+續	1
+雙	1
+僅	1
+餘	1
+員	1
+層	1
+佔	1
+據	1
+優	1
+擁	1
+穫	1
+問	1
+題	1
+陸	1
+舀	1
+毽	1
+牒	1
+缫	1
+倬	1
+庶	1
+噔	1
+锨	1
+塄	1
+碛	1
+铱	1
+囹	1
+圄	1
+廭	1
+滪	1
+戾	1
+屨	1
+羸	1
+鹺	1
+溫	1
+峬	1
+鲲	1
+锆	1
+珽	1
+掸	1
+洎	1
+潠	1
+浍	1
+铛	1
+谏	1
+幛	1
+邙	1
+砦	1
+夼	1
+咛	1
+拚	1
+峄	1
+絾	1
+鴴	1
+虛	1
+絳	1
+踾	1
+尲	1
+鸴	1
+珺	1
+汴	1
+篁	1
+贇	1
+畲	1
+迤	1
+逦	1
+岼	1
+輰	1
+豾	1
+緸	1
+鶨	1
+蓼	1
+屻	1
+弲	1
+噼	1
+痈	1
+豢	1
+藿	1
+矸	1
+幇	1
+殑	1
+椈	1
+熼	1
+粍	1
+ㄤ	1
+笂	1
+彫	1
+拰	1
+伅	1
+儴	1
+鍓	1
+闀	1
+畫	1
+煴	1
+	1
+嬪	1
+炲	1
+壇	1
+鎬	1
+洃	1
+鍒	1
+埆	1
+呬	1
+岄	1
+洟	1
+暱	1
+	1
+	1
+	1
+屼	1
+缃	1
+戦	1
+勫	1
+垝	1
+杈	1
+尯	1
+鍏	1
+紬	1
+忚	1
+皟	1
+彿	1
+囩	1
+洊	1
+娾	1
+绁	1
+濋	1
+洖	1
+褰	1
+鍛	1
+鈥	1
+濅	1
+夎	1
+	1
+樺	1
+睍	1
+嗚	1
+	1
+樹	1
+粙	1
+嶄	1
+	1
+撶	1
+彂	1
+曟	1
+儏	1
+嬬	1
+粛	1
+嗛	1
+囪	1
+祫	1
+競	1
+寲	1
+傞	1
+繃	1
+弻	1
+鎴	1
+椤	1
+嗭	1
+	1
+傛	1
+	1
+篃	1
+嗘	1
+湁	1
+╀	1
+簬	1
+煙	1
+鐨	1
+勪	1
+氦	1
+綔	1
+垜	1
+敤	1
+╁	1
+簭	1
+琁	1
+澶	1
+槸	1
+绗	1
+嶈	1
+	1
+闂	1
+	1
+鐩	1
+祴	1
+	1
+蹇	1
+鎵	1
+寳	1
+笢	1
+戣	1
+爺	1
+墍	1
+況	1
+吡	1
+殚	1
+汚	1
+杻	1
+砗	1
+磲	1
+蕤	1
+堉	1
+璘	1
+苜	1
+蓿	1
+髌	1
+姹	1
+磴	1
+觞	1
+赭	1
+愎	1
+吠	1
+尕	1
+糌	1
+偌	1
+驮	1
+镒	1
+奘	1
+汜	1
+趵	1
+轱	1
+盂	1
+孬	1
+胄	1
+煸	1
+疽	1
+杵	1
+谗	1
+炆	1
+菡	1
+箴	1
+渑	1
+锉	1
+锜	1
+乜	1
+掼	1
+铖	1
+壵	1
+忤	1
+镗	1
+撵	1
+搐	1
+坳	1
+鞥	1
+嚟	1
+緑	1
+郗	1
+禛	1
+蝗	1
+赊	1
+羲	1
+萼	1
+捌	1
+肓	1
+桧	1
+焖	1
+驷	1
+磡	1
+嫔	1
+垠	1
+嫱	1
+嬲	1
+桼	1
+蜷	1
+闾	1
+珥	1
+炔	1
+唰	1
+龇	1
+蚬	1
+蚶	1
+茼	1
+钕	1
+砩	1
+挲	1
+鳊	1
+镡	1
+锫	1
+冽	1
+阊	1
+捱	1
+蟆	1
+蒯	1
+怵	1
+刽	1
+玹	1
+鳃	1
+犒	1
+倌	1
+蓊	1
+沔	1
+墁	1
+盁	1
+啜	1
+珅	1
+臊	1
+嬖	1
+彧	1
+炕	1
+胴	1
+岵	1
+茕	1

model_zoo/speech_model/keras_backend.py

@@ -0,0 +1,509 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+#
+# Copyright 2016-2099 Ailemon.net
+#
+# This file is part of ASRT Speech Recognition Tool.
+#
+# ASRT is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+# ASRT is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with ASRT.  If not, see <https://www.gnu.org/licenses/>.
+# ============================================================================
+
+"""
+@author: nl8590687
+若干声学模型模型的定义
+"""
+
+import tensorflow as tf
+from tensorflow.keras.models import Model
+from tensorflow.keras.layers import Dense, Dropout, Input, Reshape, BatchNormalization
+from tensorflow.keras.layers import Lambda, Activation, Conv2D, MaxPooling2D
+from tensorflow.keras import backend as K
+import numpy as np
+from utils.ops import ctc_decode_delete_tail_blank
+
+
+class BaseModel:
+    """
+    定义声学模型类型的接口基类
+    """
+    def __init__(self):
+        self.input_shape = None
+        self.output_shape = None
+        self.model = None
+        self.model_base = None
+        self._model_name = None
+
+    def get_model(self) -> tuple:
+        return self.model, self.model_base
+
+    def get_train_model(self) -> Model:
+        return self.model
+
+    def get_eval_model(self) -> Model:
+        return self.model_base
+
+    def summary(self) -> None:
+        self.model.summary()
+
+    def get_model_name(self) -> str:
+        return self._model_name
+
+    def load_weights(self, filename: str) -> None:
+        self.model.load_weights(filename)
+
+    def save_weights(self, filename: str) -> None:
+        self.model.save_weights(filename + '.model.h5')
+        self.model_base.save_weights(filename + '.model.base.h5')
+
+        f = open('epoch_'+self._model_name+'.txt', 'w')
+        f.write(filename)
+        f.close()
+
+    def get_loss_function(self):
+        raise Exception("method not implemented")
+    
+    def forward(self, x):
+        raise Exception("method not implemented")
+
+
+def ctc_lambda_func(args):
+    y_pred, labels, input_length, label_length = args
+    y_pred = y_pred[:, :, :]
+    return K.ctc_batch_cost(labels, y_pred, input_length, label_length)
+
+
+class SpeechModel251BN(BaseModel):
+    """
+    定义CNN+CTC模型，使用函数式模型
+
+    输入层：200维的特征值序列，一条语音数据的最大长度设为1600（大约16s）\\
+    隐藏层：卷积池化层，卷积核大小为3x3，池化窗口大小为2 \\
+    隐藏层：全连接层 \\
+    输出层：全连接层，神经元数量为self.MS_OUTPUT_SIZE，使用softmax作为激活函数， \\
+    CTC层：使用CTC的loss作为损失函数，实现连接性时序多输出
+
+    参数： \\
+        input_shape: tuple，默认值(1600, 200, 1) \\
+        output_shape: tuple，默认值(200, 1428)
+    """
+    def __init__(self, input_shape: tuple = (1600, 200, 1), output_size: int = 1428) -> None:
+        super().__init__()
+        self.input_shape = input_shape
+        self._pool_size = 8
+        self.output_shape = (input_shape[0] // self._pool_size, output_size)
+        self._model_name = 'SpeechModel251bn'
+        self.model, self.model_base = self._define_model(self.input_shape, self.output_shape[1])
+
+    def _define_model(self, input_shape, output_size) -> tuple:
+        label_max_string_length = 64
+
+        input_data = Input(name='the_input', shape=input_shape)
+
+        layer_h = Conv2D(32, (3, 3), use_bias=True, padding='same', kernel_initializer='he_normal', name='Conv0')(input_data)  # 卷积层
+        layer_h = BatchNormalization(epsilon=0.0002, name='BN0')(layer_h)
+        layer_h = Activation('relu', name='Act0')(layer_h)
+
+        layer_h = Conv2D(32, (3,3), use_bias=True, padding='same', kernel_initializer='he_normal', name='Conv1')(layer_h)  # 卷积层
+        layer_h = BatchNormalization(epsilon=0.0002, name='BN1')(layer_h)
+        layer_h = Activation('relu', name='Act1')(layer_h)
+
+        layer_h = MaxPooling2D(pool_size=2, strides=None, padding="valid")(layer_h) # 池化层
+
+        layer_h = Conv2D(64, (3,3), use_bias=True, padding='same', kernel_initializer='he_normal', name='Conv2')(layer_h)  # 卷积层
+        layer_h = BatchNormalization(epsilon=0.0002, name='BN2')(layer_h)
+        layer_h = Activation('relu', name='Act2')(layer_h)
+
+        layer_h = Conv2D(64, (3,3), use_bias=True, padding='same', kernel_initializer='he_normal', name='Conv3')(layer_h)  # 卷积层
+        layer_h = BatchNormalization(epsilon=0.0002, name='BN3')(layer_h)
+        layer_h = Activation('relu', name='Act3')(layer_h)
+
+        layer_h = MaxPooling2D(pool_size=2, strides=None, padding="valid")(layer_h) # 池化层
+
+        layer_h = Conv2D(128, (3,3), use_bias=True, padding='same', kernel_initializer='he_normal', name='Conv4')(layer_h)  # 卷积层
+        layer_h = BatchNormalization(epsilon=0.0002, name='BN4')(layer_h)
+        layer_h = Activation('relu', name='Act4')(layer_h)
+
+        layer_h = Conv2D(128, (3,3), use_bias=True, padding='same', kernel_initializer='he_normal', name='Conv5')(layer_h)  # 卷积层
+        layer_h = BatchNormalization(epsilon=0.0002, name='BN5')(layer_h)
+        layer_h = Activation('relu', name='Act5')(layer_h)
+
+        layer_h = MaxPooling2D(pool_size=2, strides=None, padding="valid")(layer_h) # 池化层
+
+        layer_h = Conv2D(128, (3,3), use_bias=True, padding='same', kernel_initializer='he_normal', name='Conv6')(layer_h)  # 卷积层
+        layer_h = BatchNormalization(epsilon=0.0002, name='BN6')(layer_h)
+        layer_h = Activation('relu', name='Act6')(layer_h)
+
+        layer_h = Conv2D(128, (3,3), use_bias=True, padding='same', kernel_initializer='he_normal', name='Conv7')(layer_h)  # 卷积层
+        layer_h = BatchNormalization(epsilon=0.0002, name='BN7')(layer_h)
+        layer_h = Activation('relu', name='Act7')(layer_h)
+
+        layer_h = MaxPooling2D(pool_size=1, strides=None, padding="valid")(layer_h) # 池化层
+
+        layer_h = Conv2D(128, (3,3), use_bias=True, padding='same', kernel_initializer='he_normal', name='Conv8')(layer_h)  # 卷积层
+        layer_h = BatchNormalization(epsilon=0.0002, name='BN8')(layer_h)
+        layer_h = Activation('relu', name='Act8')(layer_h)
+
+        layer_h = Conv2D(128, (3,3), use_bias=True, padding='same', kernel_initializer='he_normal', name='Conv9')(layer_h)  # 卷积层
+        layer_h = BatchNormalization(epsilon=0.0002, name='BN9')(layer_h)
+        layer_h = Activation('relu', name='Act9')(layer_h)
+
+        layer_h = MaxPooling2D(pool_size=1, strides=None, padding="valid")(layer_h) # 池化层
+
+        # test=Model(inputs = input_data, outputs = layer_h12)
+        # test.summary()
+
+        layer_h = Reshape((self.output_shape[0], input_shape[1] // self._pool_size * 128), name='Reshape0')(layer_h)  # Reshape层
+
+        layer_h = Dense(128, activation="relu", use_bias=True, kernel_initializer='he_normal', name='Dense0')(layer_h)  # 全连接层
+
+        layer_h = Dense(output_size, use_bias=True, kernel_initializer='he_normal', name='Dense1')(layer_h) # 全连接层
+        y_pred = Activation('softmax', name='Activation0')(layer_h)
+
+        model_base = Model(inputs = input_data, outputs = y_pred)
+        # model_data.summary()
+
+        labels = Input(name='the_labels', shape=[label_max_string_length], dtype='float32')
+        input_length = Input(name='input_length', shape=[1], dtype='int64')
+        label_length = Input(name='label_length', shape=[1], dtype='int64')
+        # Keras doesn't currently support loss funcs with extra parameters
+        # so CTC loss is implemented in a lambda layer
+        loss_out = Lambda(ctc_lambda_func, output_shape=(1,), name='ctc')([y_pred, labels, input_length, label_length])
+
+        model = Model(inputs=[input_data, labels, input_length, label_length], outputs=loss_out)
+
+        return model, model_base
+
+    def get_loss_function(self) -> dict:
+        return {'ctc': lambda y_true, y_pred: y_pred}
+
+    def forward(self, data_input):
+        batch_size = 1 
+        in_len = np.zeros((batch_size,), dtype=np.int32)
+
+        in_len[0] = self.output_shape[0]
+
+        x_in = np.zeros((batch_size,) + self.input_shape, dtype=np.float64)
+
+        for i in range(batch_size):
+            x_in[i, 0:len(data_input)] = data_input
+
+        base_pred = self.model_base.predict(x=x_in)
+        r = K.ctc_decode(base_pred, in_len, greedy=True, beam_width=100, top_paths=1)
+
+        if tf.__version__[0:2] == '1.':
+            r1 = r[0][0].eval(session=tf.compat.v1.Session())
+        else:
+            r1 = r[0][0].numpy()
+        
+        speech_result = ctc_decode_delete_tail_blank(r1[0])
+        return speech_result
+
+
+class SpeechModel251(BaseModel):
+    """
+    定义CNN+CTC模型，使用函数式模型
+
+    输入层：200维的特征值序列，一条语音数据的最大长度设为1600（大约16s）\\
+    隐藏层：卷积池化层，卷积核大小为3x3，池化窗口大小为2 \\
+    隐藏层：全连接层 \\
+    输出层：全连接层，神经元数量为self.MS_OUTPUT_SIZE，使用softmax作为激活函数， \\
+    CTC层：使用CTC的loss作为损失函数，实现连接性时序多输出
+
+    参数： \\
+        input_shape: tuple，默认值(1600, 200, 1) \\
+        output_shape: tuple，默认值(200, 1428)
+    """
+    def __init__(self, input_shape: tuple = (1600, 200, 1), output_size: int = 1428) -> None:
+        super().__init__()
+        self.input_shape = input_shape
+        self._pool_size = 8
+        self.output_shape = (input_shape[0] // self._pool_size, output_size)
+        self._model_name = 'SpeechModel251'
+        self.model, self.model_base = self._define_model(self.input_shape, self.output_shape[1])
+
+    def _define_model(self, input_shape, output_size) -> tuple:
+        label_max_string_length = 64
+
+        input_data = Input(name='the_input', shape=input_shape)
+
+        layer_h1 = Conv2D(32, (3,3), use_bias=False, activation='relu', padding='same', kernel_initializer='he_normal')(input_data)  # 卷积层
+        layer_h1 = Dropout(0.05)(layer_h1)
+        layer_h2 = Conv2D(32, (3,3), use_bias=True, activation='relu', padding='same', kernel_initializer='he_normal')(layer_h1)  # 卷积层
+        layer_h3 = MaxPooling2D(pool_size=2, strides=None, padding="valid")(layer_h2)  # 池化层
+        layer_h3 = Dropout(0.05)(layer_h3)  # 随机中断部分神经网络连接，防止过拟合
+
+        layer_h4 = Conv2D(64, (3,3), use_bias=True, activation='relu', padding='same', kernel_initializer='he_normal')(layer_h3)  # 卷积层
+        layer_h4 = Dropout(0.1)(layer_h4)
+        layer_h5 = Conv2D(64, (3,3), use_bias=True, activation='relu', padding='same', kernel_initializer='he_normal')(layer_h4)  # 卷积层
+        layer_h6 = MaxPooling2D(pool_size=2, strides=None, padding="valid")(layer_h5) # 池化层
+
+        layer_h6 = Dropout(0.1)(layer_h6)
+        layer_h7 = Conv2D(128, (3,3), use_bias=True, activation='relu', padding='same', kernel_initializer='he_normal')(layer_h6)  # 卷积层
+        layer_h7 = Dropout(0.15)(layer_h7)
+        layer_h8 = Conv2D(128, (3,3), use_bias=True, activation='relu', padding='same', kernel_initializer='he_normal')(layer_h7)  # 卷积层
+        layer_h9 = MaxPooling2D(pool_size=2, strides=None, padding="valid")(layer_h8) # 池化层
+
+        layer_h9 = Dropout(0.15)(layer_h9)
+        layer_h10 = Conv2D(128, (3,3), use_bias=True, activation='relu', padding='same', kernel_initializer='he_normal')(layer_h9)  # 卷积层
+        layer_h10 = Dropout(0.2)(layer_h10)
+        layer_h11 = Conv2D(128, (3,3), use_bias=True, activation='relu', padding='same', kernel_initializer='he_normal')(layer_h10)  # 卷积层
+        layer_h12 = MaxPooling2D(pool_size=1, strides=None, padding="valid")(layer_h11) # 池化层
+
+        layer_h12 = Dropout(0.2)(layer_h12)
+        layer_h13 = Conv2D(128, (3,3), use_bias=True, activation='relu', padding='same', kernel_initializer='he_normal')(layer_h12)  # 卷积层
+        layer_h13 = Dropout(0.2)(layer_h13)
+        layer_h14 = Conv2D(128, (3,3), use_bias=True, activation='relu', padding='same', kernel_initializer='he_normal')(layer_h13)  # 卷积层
+        layer_h15 = MaxPooling2D(pool_size=1, strides=None, padding="valid")(layer_h14) # 池化层
+
+        # test=Model(inputs = input_data, outputs = layer_h12)
+        # test.summary()
+
+        layer_h16 = Reshape((self.output_shape[0], input_shape[1] // self._pool_size * 128))(layer_h15)  # Reshape层
+        layer_h16 = Dropout(0.3)(layer_h16)  # 随机中断部分神经网络连接，防止过拟合
+        layer_h17 = Dense(128, activation="relu", use_bias=True, kernel_initializer='he_normal')(layer_h16)  # 全连接层
+        layer_h17 = Dropout(0.3)(layer_h17)
+        layer_h18 = Dense(output_size, use_bias=True, kernel_initializer='he_normal')(layer_h17)  # 全连接层
+        y_pred = Activation('softmax', name='Activation0')(layer_h18)
+
+        model_base = Model(inputs=input_data, outputs=y_pred)
+        # model_data.summary()
+
+        labels = Input(name='the_labels', shape=[label_max_string_length], dtype='float32')
+        input_length = Input(name='input_length', shape=[1], dtype='int64')
+        label_length = Input(name='label_length', shape=[1], dtype='int64')
+        # Keras doesn't currently support loss funcs with extra parameters
+        # so CTC loss is implemented in a lambda layer
+        loss_out = Lambda(ctc_lambda_func, output_shape=(1,), name='ctc')([y_pred, labels, input_length, label_length])
+
+        model = Model(inputs=[input_data, labels, input_length, label_length], outputs=loss_out)
+
+        return model, model_base
+
+    def get_loss_function(self) -> dict:
+        return {'ctc': lambda y_true, y_pred: y_pred}
+
+    def forward(self, data_input):
+        batch_size = 1 
+        in_len = np.zeros((batch_size,), dtype=np.int32)
+
+        in_len[0] = self.output_shape[0]
+
+        x_in = np.zeros((batch_size,) + self.input_shape, dtype=np.float64)
+
+        for i in range(batch_size):
+            x_in[i, 0:len(data_input)] = data_input
+
+        base_pred = self.model_base.predict(x = x_in)
+        r = K.ctc_decode(base_pred, in_len, greedy=True, beam_width=100, top_paths=1)
+
+        if tf.__version__[0:2] == '1.':
+            r1 = r[0][0].eval(session=tf.compat.v1.Session())
+        else:
+            r1 = r[0][0].numpy()
+        
+        speech_result = ctc_decode_delete_tail_blank(r1[0])
+        return speech_result
+
+
+class SpeechModel25(BaseModel):
+    """
+    定义CNN+CTC模型，使用函数式模型
+
+    输入层：200维的特征值序列，一条语音数据的最大长度设为1600（大约16s）\\
+    隐���层：卷积池化层，卷积核大小为3x3，池化窗口大小为2 \\
+    隐藏层：全连接层 \\
+    输出层：全连接层，神经元数量为self.MS_OUTPUT_SIZE，使用softmax作为激活函数， \\
+    CTC层：使用CTC的loss作为损失函数，实现连接性时序多输出
+
+    参数： \\
+        input_shape: tuple，默认值(1600, 200, 1) \\
+        output_shape: tuple，默认值(200, 1428)
+    """
+    def __init__(self, input_shape: tuple = (1600, 200, 1), output_size: int = 1428) -> None:
+        super().__init__()
+        self.input_shape = input_shape
+        self._pool_size = 8
+        self.output_shape = (input_shape[0] // self._pool_size, output_size)
+        self._model_name = 'SpeechModel25'
+        self.model, self.model_base = self._define_model(self.input_shape, self.output_shape[1])
+
+    def _define_model(self, input_shape, output_size) -> tuple:
+        label_max_string_length = 64
+
+        input_data = Input(name='the_input', shape=input_shape)
+
+        layer_h1 = Conv2D(32, (3, 3), use_bias=False, activation='relu', padding='same', kernel_initializer='he_normal')(input_data)  # 卷积层
+        layer_h1 = Dropout(0.05)(layer_h1)
+        layer_h2 = Conv2D(32, (3, 3), use_bias=True, activation='relu', padding='same', kernel_initializer='he_normal')(layer_h1)  # 卷积层
+        layer_h3 = MaxPooling2D(pool_size=2, strides=None, padding="valid")(layer_h2)  # 池化层
+        layer_h3 = Dropout(0.05)(layer_h3)  # 随机中断部分神经网络连接，防止过拟合
+
+        layer_h4 = Conv2D(64, (3, 3), use_bias=True, activation='relu', padding='same', kernel_initializer='he_normal')(layer_h3)  # 卷积层
+        layer_h4 = Dropout(0.1)(layer_h4)
+        layer_h5 = Conv2D(64, (3, 3), use_bias=True, activation='relu', padding='same', kernel_initializer='he_normal')(layer_h4)  # 卷积层
+        layer_h6 = MaxPooling2D(pool_size=2, strides=None, padding="valid")(layer_h5) # 池化层
+
+        layer_h6 = Dropout(0.1)(layer_h6)
+        layer_h7 = Conv2D(128, (3, 3), use_bias=True, activation='relu', padding='same', kernel_initializer='he_normal')(layer_h6)  # 卷积层
+        layer_h7 = Dropout(0.15)(layer_h7)
+        layer_h8 = Conv2D(128, (3,3), use_bias=True, activation='relu', padding='same', kernel_initializer='he_normal')(layer_h7)  # 卷积层
+        layer_h9 = MaxPooling2D(pool_size=2, strides=None, padding="valid")(layer_h8)  # 池化层
+
+        layer_h9 = Dropout(0.15)(layer_h9)
+        layer_h10 = Conv2D(128, (3, 3), use_bias=True, activation='relu', padding='same', kernel_initializer='he_normal')(layer_h9)  # 卷积层
+        layer_h10 = Dropout(0.2)(layer_h10)
+        layer_h11 = Conv2D(128, (3, 3), use_bias=True, activation='relu', padding='same', kernel_initializer='he_normal')(layer_h10)  # 卷积层
+        layer_h12 = MaxPooling2D(pool_size=1, strides=None, padding="valid")(layer_h11) # 池化层
+
+        # test=Model(inputs = input_data, outputs = layer_h12)
+        # test.summary()
+
+        layer_h12 = Reshape((self.output_shape[0], input_shape[1] // self._pool_size * 128))(layer_h12)  # Reshape层
+        layer_h12 = Dropout(0.3)(layer_h12)  # 随机中断部分神经网络连接，防止过拟合
+        layer_h13 = Dense(128, activation="relu", use_bias=True, kernel_initializer='he_normal')(layer_h12)  # 全连接层
+        layer_h13 = Dropout(0.3)(layer_h13)
+        layer_h14 = Dense(output_size, use_bias=True, kernel_initializer='he_normal')(layer_h13)  # 全连接层
+        y_pred = Activation('softmax', name='Activation0')(layer_h14)
+
+        model_base = Model(inputs=input_data, outputs=y_pred)
+        # model_data.summary()
+
+        labels = Input(name='the_labels', shape=[label_max_string_length], dtype='float32')
+        input_length = Input(name='input_length', shape=[1], dtype='int64')
+        label_length = Input(name='label_length', shape=[1], dtype='int64')
+        # Keras doesn't currently support loss funcs with extra parameters
+        # so CTC loss is implemented in a lambda layer
+        loss_out = Lambda(ctc_lambda_func, output_shape=(1,), name='ctc')([y_pred, labels, input_length, label_length])
+
+        model = Model(inputs=[input_data, labels, input_length, label_length], outputs=loss_out)
+
+        return model, model_base
+
+    def get_loss_function(self) -> dict:
+        return {'ctc': lambda y_true, y_pred: y_pred}
+
+    def forward(self, data_input):
+        batch_size = 1 
+        in_len = np.zeros((batch_size,), dtype=np.int32)
+
+        in_len[0] = self.output_shape[0]
+
+        x_in = np.zeros((batch_size,) + self.input_shape, dtype=np.float64)
+
+        for i in range(batch_size):
+            x_in[i, 0:len(data_input)] = data_input
+
+        base_pred = self.model_base.predict(x=x_in)
+        r = K.ctc_decode(base_pred, in_len, greedy=True, beam_width=100, top_paths=1)
+
+        if tf.__version__[0:2] == '1.':
+            r1 = r[0][0].eval(session=tf.compat.v1.Session())
+        else:
+            r1 = r[0][0].numpy()
+        
+        speech_result = ctc_decode_delete_tail_blank(r1[0])
+        return speech_result
+
+
+class SpeechModel24(BaseModel):
+    """
+    定义CNN+CTC模型，使用函数式模型
+
+    输入层：200维的特征值序列，一条语音数据的最大长度设为1600（大约16s）\\
+    隐藏层：卷积池化层，卷积核大小为3x3，池化窗口大小为2 \\
+    隐藏层：全连接层 \\
+    输出层：全连接层，神经元数量为self.MS_OUTPUT_SIZE，使用softmax作为激活函数， \\
+    CTC层：使用CTC的loss作为损失函数，实现连接性时序多输出
+
+    参数： \\
+        input_shape: tuple，默认值(1600, 200, 1) \\
+        output_shape: tuple，默认值(200, 1428)
+    """
+    def __init__(self, input_shape :tuple=(1600, 200, 1), output_size: int = 1428) -> None:
+        super().__init__()
+        self.input_shape = input_shape
+        self._pool_size = 8
+        self.output_shape = (input_shape[0] // self._pool_size, output_size)
+        self._model_name = 'SpeechModel24'
+        self.model, self.model_base = self._define_model(self.input_shape, self.output_shape[1])
+
+    def _define_model(self, input_shape, output_size) -> tuple:
+        label_max_string_length = 64
+
+        input_data = Input(name='the_input', shape=input_shape)
+
+        layer_h1 = Conv2D(32, (3,3), use_bias=False, activation='relu', padding='same', kernel_initializer='he_normal')(input_data)  # 卷积层
+        layer_h1 = Dropout(0.1)(layer_h1)
+        layer_h2 = Conv2D(32, (3,3), use_bias=True, activation='relu', padding='same', kernel_initializer='he_normal')(layer_h1)  # 卷积层
+        layer_h3 = MaxPooling2D(pool_size=2, strides=None, padding="valid")(layer_h2)  # 池化层
+        layer_h3 = Dropout(0.2)(layer_h3)  # 随机中断部分神经网络连接，防止过拟合
+
+        layer_h4 = Conv2D(64, (3,3), use_bias=True, activation='relu', padding='same', kernel_initializer='he_normal')(layer_h3)  # 卷积层
+        layer_h4 = Dropout(0.2)(layer_h4)
+        layer_h5 = Conv2D(64, (3,3), use_bias=True, activation='relu', padding='same', kernel_initializer='he_normal')(layer_h4)  # 卷积层
+        layer_h6 = MaxPooling2D(pool_size=2, strides=None, padding="valid")(layer_h5) # 池化层
+
+        layer_h6 = Dropout(0.3)(layer_h6)
+        layer_h7 = Conv2D(128, (3,3), use_bias=True, activation='relu', padding='same', kernel_initializer='he_normal')(layer_h6)  # 卷积层
+        layer_h7 = Dropout(0.3)(layer_h7)
+        layer_h8 = Conv2D(128, (3,3), use_bias=True, activation='relu', padding='same', kernel_initializer='he_normal')(layer_h7)  # 卷积层
+        layer_h9 = MaxPooling2D(pool_size=2, strides=None, padding="valid")(layer_h8) # 池化层
+
+        # test=Model(inputs = input_data, outputs = layer_h12)
+        # test.summary()
+
+        layer_h10 = Reshape((self.output_shape[0], input_shape[1] // self._pool_size * 128))(layer_h9)  # Reshape层
+        layer_h10 = Dropout(0.3)(layer_h10)  # 随机中断部分神经网络连接，防止过拟合
+        layer_h11 = Dense(128, activation="relu", use_bias=True, kernel_initializer='he_normal')(layer_h10) # 全连接层
+        layer_h11 = Dropout(0.3)(layer_h11)
+        layer_h12 = Dense(output_size, use_bias=True, kernel_initializer='he_normal')(layer_h11) # 全连接层
+        y_pred = Activation('softmax', name='Activation0')(layer_h12)
+
+        model_base = Model(inputs=input_data, outputs=y_pred)
+        # model_data.summary()
+
+        labels = Input(name='the_labels', shape=[label_max_string_length], dtype='float32')
+        input_length = Input(name='input_length', shape=[1], dtype='int64')
+        label_length = Input(name='label_length', shape=[1], dtype='int64')
+        # Keras doesn't currently support loss funcs with extra parameters
+        # so CTC loss is implemented in a lambda layer
+        loss_out = Lambda(ctc_lambda_func, output_shape=(1,), name='ctc')([y_pred, labels, input_length, label_length])
+
+        model = Model(inputs=[input_data, labels, input_length, label_length], outputs=loss_out)
+
+        return model, model_base
+
+    def get_loss_function(self) -> dict:
+        return {'ctc': lambda y_true, y_pred: y_pred}
+
+    def forward(self, data_input):
+        batch_size = 1 
+        in_len = np.zeros((batch_size,), dtype=np.int32)
+
+        in_len[0] = self.output_shape[0]
+
+        x_in = np.zeros((batch_size,) + self.input_shape, dtype=np.float64)
+
+        for i in range(batch_size):
+            x_in[i, 0:len(data_input)] = data_input
+
+        base_pred = self.model_base.predict(x=x_in)
+        r = K.ctc_decode(base_pred, in_len, greedy=True, beam_width=100, top_paths=1)
+
+        if tf.__version__[0:2] == '1.':
+            r1 = r[0][0].eval(session=tf.compat.v1.Session())
+        else:
+            r1 = r[0][0].numpy()
+        
+        speech_result = ctc_decode_delete_tail_blank(r1[0])
+        return speech_result

predict_speech_file.py

@@ -0,0 +1,63 @@
+'''
+Author: wuxulong19950206 [email protected]
+Date: 2024-03-09 15:53:33
+LastEditors: wuxulong19950206 [email protected]
+LastEditTime: 2024-03-10 00:00:12
+FilePath: \ASRT_SpeechRecognition\predict_speech_file.py
+Description: 这是默认设置,请设置`customMade`, 打开koroFileHeader查看配置 进行设置: https://github.com/OBKoro1/koro1FileHeader/wiki/%E9%85%8D%E7%BD%AE
+'''
+# !/usr/bin/env python3
+# -*- coding: utf-8 -*-
+#
+# Copyright 2016-2099 Ailemon.net
+#
+# This file is part of ASRT Speech Recognition Tool.
+#
+# ASRT is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+# ASRT is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with ASRT.  If not, see <https://www.gnu.org/licenses/>.
+# ============================================================================
+
+"""
+@author: nl8590687
+用于通过ASRT语音识别系统预测一次语音文件的程序
+"""
+
+import os
+
+from speech_model import ModelSpeech
+from model_zoo.speech_model.keras_backend import SpeechModel251BN
+from speech_features import Spectrogram
+from language_model3 import ModelLanguage
+
+os.environ["CUDA_VISIBLE_DEVICES"] = "0"
+
+AUDIO_LENGTH = 1600
+AUDIO_FEATURE_LENGTH = 200
+CHANNELS = 1
+# 默认输出的拼音的表示大小是1428，即1427个拼音+1个空白块
+OUTPUT_SIZE = 1428
+sm251bn = SpeechModel251BN(
+    input_shape=(AUDIO_LENGTH, AUDIO_FEATURE_LENGTH, CHANNELS),
+    output_size=OUTPUT_SIZE
+)
+feat = Spectrogram()
+ms = ModelSpeech(sm251bn, feat, max_label_length=64)
+
+ms.load_model('save_models/SpeechModel251bn/' + sm251bn.get_model_name() + '.model.h5')
+res = ms.recognize_speech_from_file('A2_0.wav')
+print('*[提示] 声学模型语音识别结果：\n', res)
+
+ml = ModelLanguage('model_language')
+ml.load_model()
+str_pinyin = res
+res = ml.pinyin_to_text(str_pinyin)
+print('语音识别最终结果：\n', res)

requirements.txt

@@ -0,0 +1,59 @@
+absl-py==1.4.0
+astunparse==1.6.3
+cachetools==5.3.0
+certifi==2022.12.7
+charset-normalizer==3.0.1
+click==8.1.3
+colorama==0.4.6
+contourpy==1.0.7
+cycler==0.11.0
+Flask==2.2.2
+flatbuffers==23.1.21
+fonttools==4.38.0
+gast==0.5.3
+google-auth==2.16.0
+google-auth-oauthlib==0.4.6
+google-pasta==0.2.0
+grpcio==1.51.1
+h5py==3.8.0
+idna==3.4
+importlib-metadata==6.0.0
+itsdangerous==2.1.2
+Jinja2==3.1.2
+keras==2.8.0
+Keras-Preprocessing==1.1.2
+kiwisolver==1.4.4
+libclang==15.0.6.1
+Markdown==3.4.1
+MarkupSafe==2.1.2
+matplotlib==3.6.3
+numpy==1.24.1
+oauthlib==3.2.2
+opt-einsum==3.3.0
+packaging==23.0
+Pillow==9.4.0
+protobuf==3.19.6
+pyasn1==0.4.8
+pyasn1-modules==0.2.8
+pyparsing==3.0.9
+python-dateutil==2.8.2
+requests==2.28.2
+requests-oauthlib==1.3.1
+rsa==4.9
+scipy==1.10.0
+six==1.16.0
+tensorboard==2.8.0
+tensorboard-data-server==0.6.1
+tensorboard-plugin-wit==1.8.1
+tensorflow-estimator==2.8.0
+tensorflow-gpu==2.8.4
+tensorflow-io-gcs-filesystem==0.30.0
+termcolor==2.2.0
+typing_extensions==4.4.0
+urllib3==1.26.14
+waitress==2.1.2
+Wave==0.0.2
+Werkzeug==2.2.2
+wincertstore==0.2
+wrapt==1.14.1
+zipp==3.12.0

speech_features/__init__.py

@@ -0,0 +1,26 @@
+# !/usr/bin/env python3
+# -*- coding: utf-8 -*-
+#
+# Copyright 2016-2099 Ailemon.net
+#
+# This file is part of ASRT Speech Recognition Tool.
+#
+# ASRT is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+# ASRT is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with ASRT.  If not, see <https://www.gnu.org/licenses/>.
+# ============================================================================
+
+"""
+@author: nl8590687
+ASRT语音识别声学特征模块
+"""
+
+from .speech_features import *

speech_features/base.py

@@ -0,0 +1,238 @@
+# !/usr/bin/env python3
+# -*- coding: utf-8 -*-
+#
+# Copyright 2016-2099 Ailemon.net
+#
+# This file is part of ASRT Speech Recognition Tool.
+#
+# ASRT is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+# ASRT is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with ASRT.  If not, see <https://www.gnu.org/licenses/>.
+# ============================================================================
+
+# calculate filterbank features. Provides e.g. fbank and mfcc features for use in ASR applications
+# Author: James Lyons 2012
+
+"""
+@author: nl8590687
+ASRT语音识别声学特征基础库模块，一些基础函数实现
+"""
+
+from __future__ import division
+import numpy
+from scipy.fftpack import dct
+
+from .sigproc import preemphasis, framesig, powspec
+
+
+def calculate_nfft(samplerate, winlen):
+    """Calculates the FFT size as a power of two greater than or equal to
+    the number of samples in a single window length.
+
+    Having an FFT less than the window length loses precision by dropping
+    many of the samples; a longer FFT than the window allows zero-padding
+    of the FFT buffer which is neutral in terms of frequency domain conversion.
+    :param samplerate: The sample rate of the signal we are working with, in Hz.
+    :param winlen: The length of the analysis window in seconds.
+    """
+    window_length_samples = winlen * samplerate
+    nfft = 1
+    while nfft < window_length_samples:
+        nfft *= 2
+    return nfft
+
+
+def mfcc(signal, samplerate=16000, winlen=0.025, winstep=0.01, numcep=13,
+         nfilt=26, nfft=None, lowfreq=0, highfreq=None, preemph=0.97, ceplifter=22, appendEnergy=True,
+         winfunc=lambda x: numpy.ones((x,))):
+    """Compute MFCC features from an audio signal.
+    :param signal: the audio signal from which to compute features. Should be an N*1 array
+    :param samplerate: the sample rate of the signal we are working with, in Hz.
+    :param winlen: the length of the analysis window in seconds. Default is 0.025s (25 milliseconds)
+    :param winstep: the step between successive windows in seconds. Default is 0.01s (10 milliseconds)
+    :param numcep: the number of cepstrum to return, default 13
+    :param nfilt: the number of filters in the filterbank, default 26.
+    :param nfft: the FFT size. Default is None, which uses the calculate_nfft function to choose the smallest size that does not drop sample data.
+    :param lowfreq: lowest band edge of mel filters. In Hz, default is 0.
+    :param highfreq: highest band edge of mel filters. In Hz, default is samplerate/2
+    :param preemph: apply preemphasis filter with preemph as coefficient. 0 is no filter. Default is 0.97.
+    :param ceplifter: apply a lifter to final cepstral coefficients. 0 is no lifter. Default is 22.
+    :param appendEnergy: if this is true, the zeroth cepstral coefficient is replaced with the log of the total frame energy.
+    :param winfunc: the analysis window to apply to each frame. By default no window is applied. You can use numpy window functions here e.g. winfunc=numpy.hamming
+    :returns: A numpy array of size (NUMFRAMES by numcep) containing features. Each row holds 1 feature vector.
+    """
+    nfft = nfft or calculate_nfft(samplerate, winlen)
+    feat, energy = fbank(signal, samplerate, winlen, winstep, nfilt, nfft, lowfreq, highfreq, preemph, winfunc)
+    feat = numpy.log(feat)
+    feat = dct(feat, type=2, axis=1, norm='ortho')[:, :numcep]
+    feat = lifter(feat, ceplifter)
+    if appendEnergy: feat[:, 0] = numpy.log(energy)  # replace first cepstral coefficient with log of frame energy
+    return feat
+
+
+def fbank(signal, samplerate=16000, winlen=0.025, winstep=0.01,
+          nfilt=26, nfft=512, lowfreq=0, highfreq=None, preemph=0.97,
+          winfunc=lambda x: numpy.ones((x,))):
+    """Compute Mel-filterbank energy features from an audio signal.
+    :param signal: the audio signal from which to compute features. Should be an N*1 array
+    :param samplerate: the sample rate of the signal we are working with, in Hz.
+    :param winlen: the length of the analysis window in seconds. Default is 0.025s (25 milliseconds)
+    :param winstep: the step between successive windows in seconds. Default is 0.01s (10 milliseconds)
+    :param nfilt: the number of filters in the filterbank, default 26.
+    :param nfft: the FFT size. Default is 512.
+    :param lowfreq: lowest band edge of mel filters. In Hz, default is 0.
+    :param highfreq: highest band edge of mel filters. In Hz, default is samplerate/2
+    :param preemph: apply preemphasis filter with preemph as coefficient. 0 is no filter. Default is 0.97.
+    :param winfunc: the analysis window to apply to each frame. By default no window is applied. You can use numpy window functions here e.g. winfunc=numpy.hamming
+    :returns: 2 values. The first is a numpy array of size (NUMFRAMES by nfilt) containing features. Each row holds 1 feature vector. The
+        second return value is the energy in each frame (total energy, unwindowed)
+    """
+    highfreq = highfreq or samplerate / 2
+    signal = preemphasis(signal, preemph)
+    frames = framesig(signal, winlen * samplerate, winstep * samplerate, winfunc)
+    pspec = powspec(frames, nfft)
+    energy = numpy.sum(pspec, 1)  # this stores the total energy in each frame
+    energy = numpy.where(energy == 0, numpy.finfo(float).eps, energy)  # if energy is zero, we get problems with log
+
+    fb = get_filterbanks(nfilt, nfft, samplerate, lowfreq, highfreq)
+    feat = numpy.dot(pspec, fb.T)  # compute the filterbank energies
+    feat = numpy.where(feat == 0, numpy.finfo(float).eps, feat)  # if feat is zero, we get problems with log
+
+    return feat, energy
+
+
+def logfbank(signal, samplerate=16000, winlen=0.025, winstep=0.01,
+             nfilt=26, nfft=512, lowfreq=0, highfreq=None, preemph=0.97,
+             winfunc=lambda x: numpy.ones((x,))):
+    """Compute log Mel-filterbank energy features from an audio signal.
+    :param signal: the audio signal from which to compute features. Should be an N*1 array
+    :param samplerate: the sample rate of the signal we are working with, in Hz.
+    :param winlen: the length of the analysis window in seconds. Default is 0.025s (25 milliseconds)
+    :param winstep: the step between successive windows in seconds. Default is 0.01s (10 milliseconds)
+    :param nfilt: the number of filters in the filterbank, default 26.
+    :param nfft: the FFT size. Default is 512.
+    :param lowfreq: lowest band edge of mel filters. In Hz, default is 0.
+    :param highfreq: highest band edge of mel filters. In Hz, default is samplerate/2
+    :param preemph: apply preemphasis filter with preemph as coefficient. 0 is no filter. Default is 0.97.
+    :param winfunc: the analysis window to apply to each frame. By default no window is applied. You can use numpy window functions here e.g. winfunc=numpy.hamming
+    :returns: A numpy array of size (NUMFRAMES by nfilt) containing features. Each row holds 1 feature vector.
+    """
+    feat, energy = fbank(signal, samplerate, winlen, winstep, nfilt, nfft, lowfreq, highfreq, preemph, winfunc)
+    return numpy.log(feat)
+
+
+def ssc(signal, samplerate=16000, winlen=0.025, winstep=0.01,
+        nfilt=26, nfft=512, lowfreq=0, highfreq=None, preemph=0.97,
+        winfunc=lambda x: numpy.ones((x,))):
+    """Compute Spectral Subband Centroid features from an audio signal.
+    :param signal: the audio signal from which to compute features. Should be an N*1 array
+    :param samplerate: the sample rate of the signal we are working with, in Hz.
+    :param winlen: the length of the analysis window in seconds. Default is 0.025s (25 milliseconds)
+    :param winstep: the step between successive windows in seconds. Default is 0.01s (10 milliseconds)
+    :param nfilt: the number of filters in the filterbank, default 26.
+    :param nfft: the FFT size. Default is 512.
+    :param lowfreq: lowest band edge of mel filters. In Hz, default is 0.
+    :param highfreq: highest band edge of mel filters. In Hz, default is samplerate/2
+    :param preemph: apply preemphasis filter with preemph as coefficient. 0 is no filter. Default is 0.97.
+    :param winfunc: the analysis window to apply to each frame. By default no window is applied. You can use numpy window functions here e.g. winfunc=numpy.hamming
+    :returns: A numpy array of size (NUMFRAMES by nfilt) containing features. Each row holds 1 feature vector.
+    """
+    highfreq = highfreq or samplerate / 2
+    signal = preemphasis(signal, preemph)
+    frames = framesig(signal, winlen * samplerate, winstep * samplerate, winfunc)
+    pspec = powspec(frames, nfft)
+    pspec = numpy.where(pspec == 0, numpy.finfo(float).eps, pspec)  # if things are all zeros we get problems
+
+    fb = get_filterbanks(nfilt, nfft, samplerate, lowfreq, highfreq)
+    feat = numpy.dot(pspec, fb.T)  # compute the filterbank energies
+    R = numpy.tile(numpy.linspace(1, samplerate / 2, numpy.size(pspec, 1)), (numpy.size(pspec, 0), 1))
+
+    return numpy.dot(pspec * R, fb.T) / feat
+
+
+def hz2mel(hz):
+    """Convert a value in Hertz to Mels
+    :param hz: a value in Hz. This can also be a numpy array, conversion proceeds element-wise.
+    :returns: a value in Mels. If an array was passed in, an identical sized array is returned.
+    """
+    return 2595 * numpy.log10(1 + hz / 700.)
+
+
+def mel2hz(mel):
+    """Convert a value in Mels to Hertz
+    :param mel: a value in Mels. This can also be a numpy array, conversion proceeds element-wise.
+    :returns: a value in Hertz. If an array was passed in, an identical sized array is returned.
+    """
+    return 700 * (10 ** (mel / 2595.0) - 1)
+
+
+def get_filterbanks(nfilt=20, nfft=512, samplerate=16000, lowfreq=0, highfreq=None):
+    """Compute a Mel-filterbank. The filters are stored in the rows, the columns correspond
+    to fft bins. The filters are returned as an array of size nfilt * (nfft/2 + 1)
+    :param nfilt: the number of filters in the filterbank, default 20.
+    :param nfft: the FFT size. Default is 512.
+    :param samplerate: the sample rate of the signal we are working with, in Hz. Affects mel spacing.
+    :param lowfreq: lowest band edge of mel filters, default 0 Hz
+    :param highfreq: highest band edge of mel filters, default samplerate/2
+    :returns: A numpy array of size nfilt * (nfft/2 + 1) containing filterbank. Each row holds 1 filter.
+    """
+    highfreq = highfreq or samplerate / 2
+    assert highfreq <= samplerate / 2, "highfreq is greater than samplerate/2"
+
+    # compute points evenly spaced in mels
+    lowmel = hz2mel(lowfreq)
+    highmel = hz2mel(highfreq)
+    melpoints = numpy.linspace(lowmel, highmel, nfilt + 2)
+    # our points are in Hz, but we use fft bins, so we have to convert
+    #  from Hz to fft bin number
+    bin = numpy.floor((nfft + 1) * mel2hz(melpoints) / samplerate)
+
+    fbank = numpy.zeros([nfilt, nfft // 2 + 1])
+    for j in range(0, nfilt):
+        for i in range(int(bin[j]), int(bin[j + 1])):
+            fbank[j, i] = (i - bin[j]) / (bin[j + 1] - bin[j])
+        for i in range(int(bin[j + 1]), int(bin[j + 2])):
+            fbank[j, i] = (bin[j + 2] - i) / (bin[j + 2] - bin[j + 1])
+    return fbank
+
+
+def lifter(cepstra, L=22):
+    """Apply a cepstral lifter the the matrix of cepstra. This has the effect of increasing the
+    magnitude of the high frequency DCT coeffs.
+    :param cepstra: the matrix of mel-cepstra, will be numframes * numcep in size.
+    :param L: the liftering coefficient to use. Default is 22. L <= 0 disables lifter.
+    """
+    if L > 0:
+        nframes, ncoeff = numpy.shape(cepstra)
+        n = numpy.arange(ncoeff)
+        lift = 1 + (L / 2.) * numpy.sin(numpy.pi * n / L)
+        return lift * cepstra
+    else:
+        # values of L <= 0, do nothing
+        return cepstra
+
+
+def delta(feat, N):
+    """Compute delta features from a feature vector sequence.
+    :param feat: A numpy array of size (NUMFRAMES by number of features) containing features. Each row holds 1 feature vector.
+    :param N: For each frame, calculate delta features based on preceding and following N frames
+    :returns: A numpy array of size (NUMFRAMES by number of features) containing delta features. Each row holds 1 delta feature vector.
+    """
+    if N < 1:
+        raise ValueError('N must be an integer >= 1')
+    NUMFRAMES = len(feat)
+    denominator = 2 * sum([i ** 2 for i in range(1, N + 1)])
+    delta_feat = numpy.empty_like(feat)
+    padded = numpy.pad(feat, ((N, N), (0, 0)), mode='edge')  # padded version of feat
+    for t in range(NUMFRAMES):
+        delta_feat[t] = numpy.dot(numpy.arange(-N, N + 1),
+                                  padded[t: t + 2 * N + 1]) / denominator  # [t : t+2*N+1] == [(N+t)-N : (N+t)+N+1]
+    return delta_feat

speech_features/sigproc.py

@@ -0,0 +1,159 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+#
+# Copyright 2016-2099 Ailemon.net
+#
+# This file is part of ASRT Speech Recognition Tool.
+#
+# ASRT is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+# ASRT is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with ASRT.  If not, see <https://www.gnu.org/licenses/>.
+# ============================================================================
+
+# This file includes routines for basic signal processing including framing and computing power spectra.
+# Author: James Lyons 2012
+
+'''
+@author: nl8590687
+ASRT语音识别声学特征计算的信号处理计算的函数库
+'''
+
+import decimal
+import logging
+import math
+import numpy
+
+
+def round_half_up(number):
+    return int(decimal.Decimal(number).quantize(decimal.Decimal('1'), rounding=decimal.ROUND_HALF_UP))
+
+
+def rolling_window(a, window, step=1):
+    # http://ellisvalentiner.com/post/2017-03-21-np-strides-trick
+    shape = a.shape[:-1] + (a.shape[-1] - window + 1, window)
+    strides = a.strides + (a.strides[-1],)
+    return numpy.lib.stride_tricks.as_strided(a, shape=shape, strides=strides)[::step]
+
+
+def framesig(sig, frame_len, frame_step, winfunc=lambda x: numpy.ones((x,)), stride_trick=True):
+    """Frame a signal into overlapping frames.
+    :param sig: the audio signal to frame.
+    :param frame_len: length of each frame measured in samples.
+    :param frame_step: number of samples after the start of the previous frame that the next frame should begin.
+    :param winfunc: the analysis window to apply to each frame. By default no window is applied.
+    :param stride_trick: use stride trick to compute the rolling window and window multiplication faster
+    :returns: an array of frames. Size is NUMFRAMES by frame_len.
+    """
+    slen = len(sig)
+    frame_len = int(round_half_up(frame_len))
+    frame_step = int(round_half_up(frame_step))
+    if slen <= frame_len:
+        numframes = 1
+    else:
+        numframes = 1 + int(math.ceil((1.0 * slen - frame_len) / frame_step))
+
+    padlen = int((numframes - 1) * frame_step + frame_len)
+
+    zeros = numpy.zeros((padlen - slen,))
+    padsignal = numpy.concatenate((sig, zeros))
+    if stride_trick:
+        win = winfunc(frame_len)
+        frames = rolling_window(padsignal, window=frame_len, step=frame_step)
+    else:
+        indices = numpy.tile(numpy.arange(0, frame_len), (numframes, 1)) + numpy.tile(
+            numpy.arange(0, numframes * frame_step, frame_step), (frame_len, 1)).T
+        indices = numpy.array(indices, dtype=numpy.int32)
+        frames = padsignal[indices]
+        win = numpy.tile(winfunc(frame_len), (numframes, 1))
+
+    return frames * win
+
+
+def deframesig(frames, siglen, frame_len, frame_step, winfunc=lambda x: numpy.ones((x,))):
+    """Does overlap-add procedure to undo the action of framesig.
+    :param frames: the array of frames.
+    :param siglen: the length of the desired signal, use 0 if unknown. Output will be truncated to siglen samples.
+    :param frame_len: length of each frame measured in samples.
+    :param frame_step: number of samples after the start of the previous frame that the next frame should begin.
+    :param winfunc: the analysis window to apply to each frame. By default no window is applied.
+    :returns: a 1-D signal.
+    """
+    frame_len = round_half_up(frame_len)
+    frame_step = round_half_up(frame_step)
+    numframes = numpy.shape(frames)[0]
+    assert numpy.shape(frames)[1] == frame_len, '"frames" matrix is wrong size, 2nd dim is not equal to frame_len'
+
+    indices = numpy.tile(numpy.arange(0, frame_len), (numframes, 1)) + numpy.tile(
+        numpy.arange(0, numframes * frame_step, frame_step), (frame_len, 1)).T
+    indices = numpy.array(indices, dtype=numpy.int32)
+    padlen = (numframes - 1) * frame_step + frame_len
+
+    if siglen <= 0: siglen = padlen
+
+    rec_signal = numpy.zeros((padlen,))
+    window_correction = numpy.zeros((padlen,))
+    win = winfunc(frame_len)
+
+    for i in range(0, numframes):
+        window_correction[indices[i, :]] = window_correction[
+                                               indices[i, :]] + win + 1e-15  # add a little bit so it is never zero
+        rec_signal[indices[i, :]] = rec_signal[indices[i, :]] + frames[i, :]
+
+    rec_signal = rec_signal / window_correction
+    return rec_signal[0:siglen]
+
+
+def magspec(frames, NFFT):
+    """Compute the magnitude spectrum of each frame in frames. If frames is an NxD matrix, output will be Nx(NFFT/2+1).
+    :param frames: the array of frames. Each row is a frame.
+    :param NFFT: the FFT length to use. If NFFT > frame_len, the frames are zero-padded.
+    :returns: If frames is an NxD matrix, output will be Nx(NFFT/2+1). Each row will be the magnitude spectrum of the corresponding frame.
+    """
+    if numpy.shape(frames)[1] > NFFT:
+        logging.warn(
+            'frame length (%d) is greater than FFT size (%d), frame will be truncated. Increase NFFT to avoid.',
+            numpy.shape(frames)[1], NFFT)
+    complex_spec = numpy.fft.rfft(frames, NFFT)
+    return numpy.absolute(complex_spec)
+
+
+def powspec(frames, NFFT):
+    """Compute the power spectrum of each frame in frames. If frames is an NxD matrix, output will be Nx(NFFT/2+1).
+    :param frames: the array of frames. Each row is a frame.
+    :param NFFT: the FFT length to use. If NFFT > frame_len, the frames are zero-padded.
+    :returns: If frames is an NxD matrix, output will be Nx(NFFT/2+1). Each row will be the power spectrum of the corresponding frame.
+    """
+    return 1.0 / NFFT * numpy.square(magspec(frames, NFFT))
+
+
+def logpowspec(frames, NFFT, norm=1):
+    """Compute the log power spectrum of each frame in frames. If frames is an NxD matrix, output will be Nx(NFFT/2+1).
+    :param frames: the array of frames. Each row is a frame.
+    :param NFFT: the FFT length to use. If NFFT > frame_len, the frames are zero-padded.
+    :param norm: If norm=1, the log power spectrum is normalised so that the max value (across all frames) is 0.
+    :returns: If frames is an NxD matrix, output will be Nx(NFFT/2+1). Each row will be the log power spectrum of the corresponding frame.
+    """
+    ps = powspec(frames, NFFT)
+    ps[ps <= 1e-30] = 1e-30
+    lps = 10 * numpy.log10(ps)
+    if norm:
+        return lps - numpy.max(lps)
+    else:
+        return lps
+
+
+def preemphasis(signal, coeff=0.95):
+    """perform preemphasis on the input signal.
+    :param signal: The signal to filter.
+    :param coeff: The preemphasis coefficient. 0 is no filter, default is 0.95.
+    :returns: the filtered signal.
+    """
+    return numpy.append(signal[0], signal[1:] - coeff * signal[:-1])

speech_features/speech_features.py

@@ -0,0 +1,228 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+#
+# Copyright 2016-2099 Ailemon.net
+#
+# This file is part of ASRT Speech Recognition Tool.
+#
+# ASRT is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+# ASRT is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with ASRT.  If not, see <https://www.gnu.org/licenses/>.
+# ============================================================================
+
+"""
+@author: nl8590687
+ASRT语音识别内置声学特征提取模块，定义了几个常用的声学特征类
+"""
+
+import random
+import numpy as np
+from scipy.fftpack import fft
+from .base import mfcc, delta, logfbank
+
+
+class SpeechFeatureMeta:
+    """
+    ASRT语音识别中所有声学特征提取类的基类
+    """
+
+    def __init__(self, framesamplerate=16000):
+        self.framesamplerate = framesamplerate
+
+    def run(self, wavsignal, fs=16000):
+        '''
+        run method
+        '''
+        raise NotImplementedError('[ASRT] `run()` method is not implemented.')
+
+
+class MFCC(SpeechFeatureMeta):
+    """
+    ASRT语音识别内置的mfcc声学特征提取类
+
+    Compute MFCC features from an audio signal.
+
+    :param framesamplerate: the sample rate of the signal we are working with, in Hz.
+    :param winlen: the length of the analysis window in seconds. Default is 0.025s (25 milliseconds)
+    :param winstep: the step between successive windows in seconds. Default is 0.01s (10 milliseconds)
+    :param numcep: the number of cepstrum to return, default 13
+    :param nfilt: the number of filters in the filterbank, default 26.
+    :param preemph: apply preemphasis filter with preemph as coefficient. 0 is no filter. Default is 0.97.
+    """
+
+    def __init__(self, framesamplerate=16000,
+                 winlen=0.025,
+                 winstep=0.01,
+                 numcep=13,
+                 nfilt=26,
+                 preemph=0.97):
+        self.framesamplerate = framesamplerate
+        self.winlen = winlen
+        self.winstep = winstep
+        self.numcep = numcep
+        self.nfilt = nfilt
+        self.preemph = preemph
+        super().__init__(framesamplerate)
+
+    def run(self, wavsignal, fs=16000):
+        """
+        计算mfcc声学特征，包含静态特征、一阶差分和二阶差分
+
+        :returns: A numpy array of size (NUMFRAMES by numcep * 3) containing features. Each row holds 1 feature vector.
+        """
+        wavsignal = np.array(wavsignal, dtype=np.float64)
+        # 获取输入特征
+        feat_mfcc = mfcc(wavsignal[0], samplerate=self.framesamplerate, winlen=self.winlen,
+                         winstep=self.winstep, numcep=self.numcep, nfilt=self.nfilt, preemph=self.preemph)
+        feat_mfcc_d = delta(feat_mfcc, 2)
+        feat_mfcc_dd = delta(feat_mfcc_d, 2)
+        # 返回值分别是mfcc特征向量的矩阵及其一阶差分和二阶差分矩阵
+        wav_feature = np.column_stack((feat_mfcc, feat_mfcc_d, feat_mfcc_dd))
+        return wav_feature
+
+
+class Logfbank(SpeechFeatureMeta):
+    """
+    ASRT语音识别内置的logfbank声学特征提取类
+    """
+
+    def __init__(self, framesamplerate=16000, nfilt=26):
+        self.nfilt = nfilt
+        super().__init__(framesamplerate)
+
+    def run(self, wavsignal, fs=16000):
+        wavsignal = np.array(wavsignal, dtype=np.float64)
+        # 获取输入特征
+        wav_feature = logfbank(wavsignal, fs, nfilt=self.nfilt)
+        return wav_feature
+
+
+class Spectrogram(SpeechFeatureMeta):
+    """
+    ASRT语音识别内置的语谱图声学特征提取类
+    """
+
+    def __init__(self, framesamplerate=16000, timewindow=25, timeshift=10):
+        self.time_window = timewindow
+        self.window_length = int(framesamplerate / 1000 * self.time_window)  # 计算窗长度的公式，目前全部为400固定值
+        self.timeshift = timeshift
+
+        '''
+        # 保留将来用于不同采样频率
+        self.x=np.linspace(0, self.window_length - 1, self.window_length, dtype = np.int64)
+        self.w = 0.54 - 0.46 * np.cos(2 * np.pi * (self.x) / (self.window_length - 1) ) # 汉明窗
+        '''
+
+        self.x = np.linspace(0, 400 - 1, 400, dtype=np.int64)
+        self.w = 0.54 - 0.46 * np.cos(2 * np.pi * (self.x) / (400 - 1))  # 汉明窗
+        super().__init__(framesamplerate)
+
+    def run(self, wavsignal, fs=16000):
+        if fs != 16000:
+            raise ValueError(
+                f"[Error] ASRT currently only supports wav audio files with a sampling rate of 16000 Hz, but this "
+                f"audio is {fs} Hz.")
+
+        # wav波形 加时间窗以及���移10ms
+        time_window = 25  # 单位ms
+        window_length = int(fs / 1000 * time_window)  # 计算窗长度的公式，目前全部为400固定值
+
+        wav_arr = np.array(wavsignal)
+        # wav_length = len(wavsignal[0])
+        # wav_length = wav_arr.shape[1]
+
+        range0_end = int(len(wavsignal[0]) / fs * 1000 - time_window) // 10 + 1  # 计算循环终止的位置，也就是最终生成的窗数
+        data_input = np.zeros((range0_end, window_length // 2), dtype=np.float64)  # 用于存放最终的频率特征数据
+        data_line = np.zeros((1, window_length), dtype=np.float64)
+
+        for i in range(0, range0_end):
+            p_start = i * 160
+            p_end = p_start + 400
+
+            data_line = wav_arr[0, p_start:p_end]
+            data_line = data_line * self.w  # 加窗
+            data_line = np.abs(fft(data_line))
+
+            data_input[i] = data_line[0: window_length // 2]  # 设置为400除以2的值（即200）是取一半数据，因为是对称的
+
+        data_input = np.log(data_input + 1)
+        return data_input
+
+
+class SpecAugment(SpeechFeatureMeta):
+    """
+    复现谷歌SpecAugment数据增强特征算法，基于Spectrogram语谱图基础特征
+    """
+
+    def __init__(self, framesamplerate=16000, timewindow=25, timeshift=10):
+        self.time_window = timewindow
+        self.window_length = int(framesamplerate / 1000 * self.time_window)  # 计算窗长度的公式，目前全部为400固定值
+        self.timeshift = timeshift
+
+        '''
+        # 保留将来用于不同采样频率
+        self.x=np.linspace(0, self.window_length - 1, self.window_length, dtype = np.int64)
+        self.w = 0.54 - 0.46 * np.cos(2 * np.pi * (self.x) / (self.window_length - 1) ) # 汉明窗
+        '''
+
+        self.x = np.linspace(0, 400 - 1, 400, dtype=np.int64)
+        self.w = 0.54 - 0.46 * np.cos(2 * np.pi * (self.x) / (400 - 1))  # 汉明窗
+        super().__init__(framesamplerate)
+
+    def run(self, wavsignal, fs=16000):
+        if fs != 16000:
+            raise ValueError(
+                f"[Error] ASRT currently only supports wav audio files with a sampling rate of 16000 Hz, but this "
+                f"audio is {fs} Hz.")
+
+        # wav波形 加时间窗以及时移10ms
+        time_window = 25  # 单位ms
+        window_length = int(fs / 1000 * time_window)  # 计算窗长度的公式，目前全部为400固定值
+
+        wav_arr = np.array(wavsignal)
+        # wav_length = len(wavsignal[0])
+        # wav_length = wav_arr.shape[1]
+
+        range0_end = int(len(wavsignal[0]) / fs * 1000 - time_window) // 10 + 1  # 计算循环终止的位置，也就是最终生成的窗数
+        data_input = np.zeros((range0_end, window_length // 2), dtype=np.float64)  # 用于存放最终的频率特征数据
+        data_line = np.zeros((1, window_length), dtype=np.float64)
+
+        for i in range(0, range0_end):
+            p_start = i * 160
+            p_end = p_start + 400
+
+            data_line = wav_arr[0, p_start:p_end]
+            data_line = data_line * self.w  # 加窗
+            data_line = np.abs(fft(data_line))
+
+            data_input[i] = data_line[0: window_length // 2]  # 设置为400除以2的值（即200）是取一半数据，因为是对称的
+
+        # print(data_input.shape)
+        data_input = np.log(data_input + 1)
+
+        # 开始对得到的特征应用SpecAugment
+        mode = random.randint(1, 100)
+        h_start = random.randint(1, data_input.shape[0])
+        h_width = random.randint(1, 100)
+
+        v_start = random.randint(1, data_input.shape[1])
+        v_width = random.randint(1, 100)
+
+        if mode <= 60:  # 正常特征 60%
+            pass
+        elif 60 < mode <= 75:  # 横向遮盖 15%
+            data_input[h_start:h_start + h_width, :] = 0
+        elif 75 < mode <= 90:  # 纵向遮盖 15%
+            data_input[:, v_start:v_start + v_width] = 0
+        else:  # 两种遮盖叠加 10%
+            data_input[h_start:h_start + h_width, :v_start:v_start + v_width] = 0
+
+        return data_input

speech_model.py

@@ -0,0 +1,264 @@
+# !/usr/bin/env python3
+# -*- coding: utf-8 -*-
+#
+# Copyright 2016-2099 Ailemon.net
+#
+# This file is part of ASRT Speech Recognition Tool.
+#
+# ASRT is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+# ASRT is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with ASRT.  If not, see <https://www.gnu.org/licenses/>.
+# ============================================================================
+
+"""
+@author: nl8590687
+声学模型基础功能模板定义
+"""
+import os
+import time
+import random
+import numpy as np
+
+from utils.ops import get_edit_distance, read_wav_data
+from utils.config import load_config_file, DEFAULT_CONFIG_FILENAME, load_pinyin_dict
+from utils.thread import threadsafe_generator
+
+
+class ModelSpeech:
+    """
+    语音模型类
+
+    参数：
+        speech_model: 声学模型类型 (BaseModel类) 实例对象
+        speech_features: 声学特征类型(SpeechFeatureMeta类)实例对象
+    """
+
+    def __init__(self, speech_model, speech_features, max_label_length=64):
+        self.data_loader = None
+        self.speech_model = speech_model
+        self.trained_model, self.base_model = speech_model.get_model()
+        # self.load_model("save_models\SpeechModel251bn\SpeechModel251bn.model.h5")
+        self.speech_features = speech_features
+        self.max_label_length = max_label_length
+
+    @threadsafe_generator
+    def _data_generator(self, batch_size, data_loader):
+        """
+        数据生成器函数，用于Keras的generator_fit训练
+        batch_size: 一次产生的数据量
+        """
+        labels = np.zeros((batch_size, 1), dtype=np.float64)
+        data_count = data_loader.get_data_count()
+        index = 0
+
+        while True:
+            X = np.zeros((batch_size,) + self.speech_model.input_shape, dtype=np.float64)
+            y = np.zeros((batch_size, self.max_label_length), dtype=np.int16)
+            input_length = []
+            label_length = []
+
+            for i in range(batch_size):
+                wavdata, sample_rate, data_labels = data_loader.get_data(index)
+                data_input = self.speech_features.run(wavdata, sample_rate)
+                data_input = data_input.reshape(data_input.shape[0], data_input.shape[1], 1)
+                # 必须加上模pool_size得到的值，否则会出现inf问题，然后提示No valid path found.
+                # 但是直接加又可能会出现sequence_length <= xxx 的问题，因此不能让其超过时间序列长度的最大值，比如200
+                pool_size = self.speech_model.input_shape[0] // self.speech_model.output_shape[0]
+                inlen = min(data_input.shape[0] // pool_size + data_input.shape[0] % pool_size,
+                            self.speech_model.output_shape[0])
+                input_length.append(inlen)
+
+                X[i, 0:len(data_input)] = data_input
+                y[i, 0:len(data_labels)] = data_labels
+                label_length.append([len(data_labels)])
+
+                index = (index + 1) % data_count
+
+            label_length = np.matrix(label_length)
+            input_length = np.array([input_length]).T
+
+            yield [X, y, input_length, label_length], labels
+
+    def train_model(self, optimizer, data_loader, epochs=1, save_step=1, batch_size=16, last_epoch=0, call_back=None):
+        """
+        训练模型
+
+        参数：
+            optimizer：tensorflow.keras.optimizers 优化器实例对象
+            data_loader：数据加载器类型 (SpeechData) 实例对象
+            epochs: 迭代轮数
+            save_step: 每多少epoch保存一次模型
+            batch_size: mini batch大小
+            last_epoch: 上一次epoch的编号，可用于断点处继续训练时，epoch编号不冲突
+            call_back: keras call back函数
+        """
+        save_filename = os.path.join('save_models', self.speech_model.get_model_name(),
+                                     self.speech_model.get_model_name())
+
+        self.trained_model.compile(loss=self.speech_model.get_loss_function(), optimizer=optimizer)
+        print('[ASRT] Compiles Model Successfully.')
+
+        yielddatas = self._data_generator(batch_size, data_loader)
+
+        data_count = data_loader.get_data_count()  # 获取数据的数量
+        # 计算每一个epoch迭代的次数
+        num_iterate = data_count // batch_size
+        iter_start = last_epoch
+        iter_end = last_epoch + epochs
+        for epoch in range(iter_start, iter_end):  # 迭代轮数
+            try:
+                epoch += 1
+                print('[ASRT Training] train epoch %d/%d .' % (epoch, iter_end))
+                data_loader.shuffle()
+                self.trained_model.fit_generator(yielddatas, num_iterate, callbacks=call_back)
+            except StopIteration:
+                print('[error] generator error. please check data format.')
+                break
+
+            if epoch % save_step == 0:
+                if not os.path.exists('save_models'):  # 判断保存模型的目录是否存在
+                    os.makedirs('save_models')  # 如果不存在，就新建一个，避免之后保存模型的时候炸掉
+                if not os.path.exists(os.path.join('save_models', self.speech_model.get_model_name())):  # 判断保存模型的目录是否存在
+                    os.makedirs(
+                        os.path.join('save_models', self.speech_model.get_model_name()))  # 如果不存在，就新建一个，避免之后保存模型的时候炸掉
+
+                self.save_model(save_filename + '_epoch' + str(epoch))
+
+        print('[ASRT Info] Model training complete. ')
+
+    def load_model(self, filename):
+        """
+        加载模型参数
+        """
+        print(f"loading pre-trained model from {filename}")
+        self.speech_model.load_weights(filename)
+
+    def save_model(self, filename):
+        """
+        保存模型参数
+        """
+        self.speech_model.save_weights(filename)
+
+    def evaluate_model(self, data_loader, data_count=-1, out_report=False, show_ratio=True, show_per_step=100):
+        """
+        评估检验模型的识别效果
+        """
+        data_nums = data_loader.get_data_count()
+
+        if data_count <= 0 or data_count > data_nums:  # 当data_count为小于等于0或者大于测试数据量的值时，则使用全部数据来测试
+            data_count = data_nums
+
+        try:
+            ran_num = random.randint(0, data_nums - 1)  # 获取一个随机数
+            words_num = 0
+            word_error_num = 0
+
+            nowtime = time.strftime('%Y%m%d_%H%M%S', time.localtime(time.time()))
+            if out_report:
+                txt_obj = open('Test_Report_' + data_loader.dataset_type + '_' + nowtime + '.txt', 'w',
+                               encoding='UTF-8')  # 打开文件并读入
+                txt_obj.truncate((data_count + 1) * 300)  # 预先分配一定数量的磁盘空间，避免后期在硬盘中文件存储位置频繁移动，以防写入速度越来越慢
+                txt_obj.seek(0)  # 从文件首开始
+
+            txt = ''
+            i = 0
+            while i < data_count:
+                wavdata, fs, data_labels = data_loader.get_data((ran_num + i) % data_nums)  # 从随机数开始连续向后取一定数量数据
+                data_input = self.speech_features.run(wavdata, fs)
+                data_input = data_input.reshape(data_input.shape[0], data_input.shape[1], 1)
+                # 数据格式出错处理 开始
+                # 当输入的wav文件长度过长时自动跳过该文件，转而使用下一个wav文件来运行
+                if data_input.shape[0] > self.speech_model.input_shape[0]:
+                    print('*[Error]', 'wave data lenghth of num', (ran_num + i) % data_nums, 'is too long.',
+                          'this data\'s length is', data_input.shape[0],
+                          'expect <=', self.speech_model.input_shape[0],
+                          '\n A Exception raise when test Speech Model.')
+                    i += 1
+                    continue
+                # 数据格式出错处理 结束
+
+                pre = self.predict(data_input)
+
+                words_n = data_labels.shape[0]  # 获取每个句子的字数
+                words_num += words_n  # 把句子的总字数加上
+                edit_distance = get_edit_distance(data_labels, pre)  # 获取编辑距离
+                if edit_distance <= words_n:  # 当编辑距离小于等于句子字数时
+                    word_error_num += edit_distance  # 使用编辑距离作为错误字数
+                else:  # 否则肯定是增加了一堆乱七八糟的奇奇怪怪的字
+                    word_error_num += words_n  # 就直接加句子本来的总字数就好了
+
+                if i % show_per_step == 0 and show_ratio:
+                    print('[ASRT Info] Testing: ', i, '/', data_count)
+
+                txt = ''
+                if out_report:
+                    txt += str(i) + '\n'
+                    txt += 'True:\t' + str(data_labels) + '\n'
+                    txt += 'Pred:\t' + str(pre) + '\n'
+                    txt += '\n'
+                    txt_obj.write(txt)
+
+                i += 1
+
+            # print('*[测试结果] 语音识别 ' + str_dataset + ' 集语音单字错误率：', word_error_num / words_num * 100, '%')
+            print('*[ASRT Test Result] Speech Recognition ' + data_loader.dataset_type + ' set word error ratio: ',
+                  word_error_num / words_num * 100, '%')
+            if out_report:
+                txt = '*[ASRT Test Result] Speech Recognition ' + data_loader.dataset_type + ' set word error ratio: ' + str(
+                    word_error_num / words_num * 100) + ' %'
+                txt_obj.write(txt)
+                txt_obj.truncate()  # 去除文件末尾剩余未使用的空白存储字节
+                txt_obj.close()
+
+        except StopIteration:
+            print('[ASRT Error] Model testing raise a error. Please check data format.')
+
+    def predict(self, data_input):
+        """
+        预测结果
+
+        返回语音识别后的forward结果
+        """
+        return self.speech_model.forward(data_input)
+
+    def recognize_speech(self, wavsignal, fs):
+        """
+        最终做语音识别用的函数，识别一个wav序列的语音
+        """
+        # 获取输入特征
+        data_input = self.speech_features.run(wavsignal, fs)
+        data_input = np.array(data_input, dtype=np.float64)
+        # print(data_input,data_input.shape)
+        data_input = data_input.reshape(data_input.shape[0], data_input.shape[1], 1)
+        r1 = self.predict(data_input)
+        # 获取拼音列表
+        list_symbol_dic, _ = load_pinyin_dict(load_config_file(DEFAULT_CONFIG_FILENAME)['dict_filename'])
+
+        r_str = []
+        for i in r1:
+            r_str.append(list_symbol_dic[i])
+
+        return r_str
+
+    def recognize_speech_from_file(self, filename):
+        """
+        最终做语音识别用的函数，识别指定文件名的语音
+        """
+        wavsignal, sample_rate, _, _ = read_wav_data(filename)
+        r = self.recognize_speech(wavsignal, sample_rate)
+        return r
+
+    @property
+    def model(self):
+        """
+        返回tf.keras model
+        """
+        return self.trained_model

speech_recorder.py

@@ -0,0 +1,109 @@
+# !/usr/bin/env python3
+# -*- coding: utf-8 -*-
+#
+# Copyright 2016-2099 Ailemon.net
+#
+# This file is part of ASRT Speech Recognition Tool.
+#
+# ASRT is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+# ASRT is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with ASRT.  If not, see <https://www.gnu.org/licenses/>.
+# ============================================================================
+
+"""
+@author: nl8590687
+一个配置为可用于ASRT语音识别系统的录音程序
+"""
+
+import wave
+import pyaudio
+
+
+def record_wave(wavfile,
+                duration=10,
+                channels=1,
+                sampling_rate=16000,
+                sampling_bits=16,
+                chunk_size=1024,
+                keyboard_interrupt='keep_audio'):
+    """Record audio using the default audio device by PyAudio and Wave"""
+
+    format_ = None
+    if sampling_bits == 8:
+        format_ = pyaudio.paInt8
+    if sampling_bits == 16:
+        format_ = pyaudio.paInt16
+    elif sampling_bits == 24:
+        format_ = pyaudio.paInt24
+    elif sampling_bits == 32:
+        format_ = pyaudio.paFloat32
+    else:
+        raise ValueError('Unsupported sampling bits')
+
+    audio = pyaudio.PyAudio()
+    stream = audio.open(format=format_,
+                        channels=channels,
+                        rate=sampling_rate,
+                        input=True,
+                        frames_per_buffer=chunk_size)
+
+    frames = []
+
+    print('Start to record with {}-seconds audio\n'
+          'Type Ctrl-C to get an early stop (a shorter audio)'
+          .format(duration))
+    try:
+        for _ in range(0, int(sampling_rate / chunk_size * duration)):
+            data = stream.read(chunk_size)
+            frames.append(data)
+            print('.', end='', flush=True)
+    except KeyboardInterrupt:
+        if keyboard_interrupt == 'keep_audio':
+            used_seconds = int(len(frames) * chunk_size / sampling_rate)
+            print('\n-*- Early stop with {} seconds'.format(used_seconds))
+        else:
+            raise
+    print('\nRecording finished')
+
+    stream.stop_stream()
+    stream.close()
+    audio.terminate()
+
+    print('Convert PCM frames to WAV... ', end='')
+    wavfp = wave.open(wavfile, 'wb')
+    wavfp.setnchannels(channels)
+    wavfp.setsampwidth(audio.get_sample_size(format_))
+    wavfp.setframerate(sampling_rate)
+    wavfp.writeframes(b''.join(frames))
+    wavfp.close()
+    print('OK')
+
+
+if __name__ == "__main__":
+    from argparse import ArgumentParser, ArgumentDefaultsHelpFormatter
+
+    parser = ArgumentParser(description='Simple Wave Audio Recorder',
+                            formatter_class=ArgumentDefaultsHelpFormatter)
+    parser.add_argument('-d', '--duration', type=int,
+                        default=10, help='maximum duration in seconds')
+    parser.add_argument('-r', '--sampling-rate', type=int,
+                        default=16000, help='sampling rate')
+    parser.add_argument('-b', '--sampling-bits', type=int,
+                        default=16, choices=(8, 16, 24, 32), help='sampling bits')
+    parser.add_argument('-c', '--channels', type=int,
+                        default=1, help='audio channels')
+    parser.add_argument('output', nargs='?', default='output.wav',
+                        help='audio file to store audio stream')
+    args = parser.parse_args()
+    record_wave(args.output, duration=args.duration,
+                channels=args.channels,
+                sampling_bits=args.sampling_bits,
+                sampling_rate=args.sampling_rate)

train_speech_model.py

@@ -0,0 +1,61 @@
+'''
+Author: wuxulong19950206 [email protected]
+Date: 2024-03-09 15:53:33
+LastEditors: wuxulong19950206 [email protected]
+LastEditTime: 2024-03-10 11:03:04
+FilePath: \ASRT_SpeechRecognition\train_speech_model.py
+Description: 这是默认设置,请设置`customMade`, 打开koroFileHeader查看配置 进行设置: https://github.com/OBKoro1/koro1FileHeader/wiki/%E9%85%8D%E7%BD%AE
+'''
+# !/usr/bin/env python3
+# -*- coding: utf-8 -*-
+#
+# Copyright 2016-2099 Ailemon.net
+#
+# This file is part of ASRT Speech Recognition Tool.
+#
+# ASRT is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+# ASRT is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with ASRT.  If not, see <https://www.gnu.org/licenses/>.
+# ============================================================================
+
+"""
+@author: nl8590687
+用于训练语音识别系统语音模型的程序
+"""
+
+import os
+from tensorflow.keras.optimizers import Adam
+
+from speech_model import ModelSpeech
+from model_zoo.speech_model.keras_backend import SpeechModel251BN
+from data_loader import DataLoader
+from speech_features import SpecAugment
+
+os.environ["CUDA_VISIBLE_DEVICES"] = "0"
+
+AUDIO_LENGTH = 1600
+AUDIO_FEATURE_LENGTH = 200
+CHANNELS = 1
+# 默认输出的拼音的表示大小是1428，即1427个拼音+1个空白块
+OUTPUT_SIZE = 1428
+sm251bn = SpeechModel251BN(
+    input_shape=(AUDIO_LENGTH, AUDIO_FEATURE_LENGTH, CHANNELS),
+    output_size=OUTPUT_SIZE
+)
+feat = SpecAugment()
+train_data = DataLoader('train')
+opt = Adam(learning_rate=0.0001, beta_1=0.9, beta_2=0.999, decay=0.0, epsilon=10e-8)
+ms = ModelSpeech(sm251bn, feat, max_label_length=64)
+
+ms.load_model("save_models\SpeechModel251bn\SpeechModel251bn.model.h5")
+ms.train_model(optimizer=opt, data_loader=train_data,
+               epochs=50, save_step=1, batch_size=1, last_epoch=0)
+ms.save_model('save_models/' + sm251bn.get_model_name())

utils/config.py

@@ -0,0 +1,74 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+#
+# Copyright 2016-2099 Ailemon.net
+#
+# This file is part of ASRT Speech Recognition Tool.
+#
+# ASRT is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+# ASRT is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with ASRT.  If not, see <https://www.gnu.org/licenses/>.
+# ============================================================================
+
+"""
+@author: nl8590687
+加载ASRT配置文件相关
+"""
+
+import json
+
+DEFAULT_CONFIG_FILENAME = 'asrt_config.json'
+_config_dict = None
+_pinyin_dict = None
+_pinyin_list = None
+
+
+def load_config_file(filename: str) -> dict:
+    """
+    加载json配置文件
+
+    参数：\\
+        filename: 文件名
+
+    返回：\\
+        配置信息字典
+    """
+    global _config_dict
+    if _config_dict is not None:
+        return _config_dict
+
+    with open(filename, 'r', encoding="utf-8") as file_pointer:
+        _config_dict = json.load(file_pointer)
+    return _config_dict
+
+
+def load_pinyin_dict(filename: str) -> tuple:
+    """
+    加载拼音列表和拼音字典
+
+    拼音列表：用于下标索引转拼音 \\
+    拼音字典：用于拼音索引转下标
+    """
+    global _pinyin_list, _pinyin_dict
+    if _pinyin_dict is not None and _pinyin_list is not None:
+        return _pinyin_list, _pinyin_dict
+
+    _pinyin_list = list()
+    _pinyin_dict = dict()
+    with open(filename, 'r', encoding='utf-8') as file_pointer:
+        lines = file_pointer.read().split('\n')
+    for line in lines:
+        if len(line) == 0:
+            continue
+        tokens = line.split('\t')
+        _pinyin_list.append(tokens[0])
+        _pinyin_dict[tokens[0]] = len(_pinyin_list) - 1
+    return _pinyin_list, _pinyin_dict

utils/ops.py

@@ -0,0 +1,185 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+#
+# Copyright 2016-2099 Ailemon.net
+#
+# This file is part of ASRT Speech Recognition Tool.
+#
+# ASRT is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+# ASRT is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with ASRT.  If not, see <https://www.gnu.org/licenses/>.
+# ============================================================================
+
+"""
+@author: nl8590687
+一些常用操作函数的定义
+"""
+
+import wave
+import difflib
+import matplotlib.pyplot as plt
+import numpy as np
+
+import os
+def read_wav_data(filename: str) -> tuple:
+    """
+    读取一个wav文件，返回声音信号的时域谱矩阵和播放时间
+    """
+    # if os.path.exists(filename):
+    #     print("文件存在")
+    # else:
+    #     print("文件不存在",filename)
+    wav = wave.open(filename,"rb") # 打开一个wav格式的声音文件流
+    num_frame = wav.getnframes() # 获取帧数
+    num_channel=wav.getnchannels() # 获取声道数
+    framerate=wav.getframerate() # 获取帧速率
+    num_sample_width=wav.getsampwidth() # 获取实例的比特宽度，即每一帧的字节数
+    str_data = wav.readframes(num_frame) # 读取全部的帧
+    wav.close() # 关闭流
+    wave_data = np.fromstring(str_data, dtype = np.short) # 将声音文件数据转换为数组矩阵形式
+    wave_data.shape = -1, num_channel # 按照声道数将数组整形，单声道时候是一列数组，双声道时候是两列的矩阵
+    wave_data = wave_data.T # 将矩阵转置
+    return wave_data, framerate, num_channel, num_sample_width
+
+
+def read_wav_bytes(filename: str) -> tuple:
+    """
+    读取一个wav文件，返回声音信号的时域谱矩阵和播放时间
+    """
+    wav = wave.open(filename,"rb") # 打开一个wav格式的声音文件流
+    num_frame = wav.getnframes() # 获取帧数
+    num_channel=wav.getnchannels() # 获取声道数
+    framerate=wav.getframerate() # 获取帧速率
+    num_sample_width=wav.getsampwidth() # 获取实例的比特宽度，即每一帧的字节数
+    str_data = wav.readframes(num_frame) # 读取全部的帧
+    wav.close() # 关闭流
+    return str_data, framerate, num_channel, num_sample_width
+
+
+def get_edit_distance(str1, str2) -> int:
+    """
+    计算两个串的编辑距离，支持str和list类型
+    """
+    leven_cost = 0
+    sequence_match = difflib.SequenceMatcher(None, str1, str2)
+    for tag, index_1, index_2, index_j1, index_j2 in sequence_match.get_opcodes():
+        if tag == 'replace':
+            leven_cost += max(index_2-index_1, index_j2-index_j1)
+        elif tag == 'insert':
+            leven_cost += (index_j2-index_j1)
+        elif tag == 'delete':
+            leven_cost += (index_2-index_1)
+    return leven_cost
+
+
+def ctc_decode_delete_tail_blank(ctc_decode_list):
+    """
+    处理CTC解码后序列末尾余留的空白元素，删除掉
+    """
+    p = 0
+    while p < len(ctc_decode_list) and ctc_decode_list[p] != -1:
+        p += 1
+    return ctc_decode_list[0:p]
+
+
+def visual_1D(points_list, frequency=1):
+    """
+    可视化1D数据
+    """
+    # 首先创建绘图网格，1个子图
+    fig, ax = plt.subplots(1)
+    x = np.linspace(0, len(points_list)-1, len(points_list)) / frequency
+
+    # 在对应对象上调用 plot() 方法
+    ax.plot(x, points_list)
+    fig.show()
+
+
+def visual_2D(img):
+    """
+    可视化2D数据
+    """
+    plt.subplot(111)
+    plt.imshow(img)
+    plt.colorbar(cax=None, ax=None, shrink=0.5)
+    plt.show() 
+
+
+def decode_wav_bytes(samples_data: bytes, channels: int = 1, byte_width: int = 2) -> list:
+    """
+    解码wav格式样本点字节流，得到numpy数组
+    """
+    numpy_type = np.short
+    if byte_width == 4:
+        numpy_type = np.int
+    elif byte_width != 2:
+        raise Exception('error: unsurpport byte width `' + str(byte_width) + '`')
+    wave_data = np.fromstring(samples_data, dtype=numpy_type)  # 将声音文件数据转换为数组矩阵形式
+    wave_data.shape = -1, channels  # 按照声道数将数组整形，单声道时候是一列数组，双声道时候是两列的矩阵
+    wave_data = wave_data.T  # 将矩阵转置
+    return wave_data
+
+
+def get_symbol_dict(dict_filename):
+    """
+    读取拼音汉字的字典文件
+    返回读取后的字典
+    """
+    txt_obj = open(dict_filename, 'r', encoding='UTF-8') # 打开文件并读入
+    txt_text = txt_obj.read()
+    txt_obj.close()
+    txt_lines = txt_text.split('\n') # 文本分割
+
+    dic_symbol = {}  # 初始化符号字典
+    for i in txt_lines:
+        list_symbol = []  # 初始化符号列表
+        if i != '':
+            txt_l=i.split('\t')
+            pinyin = txt_l[0]
+            for word in txt_l[1]:
+                list_symbol.append(word)
+        dic_symbol[pinyin] = list_symbol
+
+    return dic_symbol
+
+
+def get_language_model(model_language_filename):
+    """
+    读取语言模型的文件
+    返回读取后的模型
+    """
+    txt_obj = open(model_language_filename, 'r', encoding='UTF-8')  # 打开文件并读入
+    txt_text = txt_obj.read()
+    txt_obj.close()
+    txt_lines = txt_text.split('\n')  # 文本分割
+
+    dic_model = {}  # 初始化符号字典
+    for i in txt_lines:
+        if i != '':
+            txt_l = i.split('\t')
+            if len(txt_l) == 1:
+                continue
+            dic_model[txt_l[0]] = txt_l[1]
+
+    return dic_model
+
+
+def ctc_decode_stream(tokens):
+    i = 0
+    while i < len(tokens):
+        while i+1 < len(tokens) and tokens[i] == tokens[i+1]:
+            i += 1
+        if i+1 == len(tokens) and tokens[i] != -1:
+            return tokens[0], []
+        if tokens[i] != -1:
+            return tokens[i], tokens[i+1:]
+        i += 1
+    return -1, []

utils/ops_test.py

@@ -0,0 +1,121 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+#
+# Copyright 2016-2099 Ailemon.net
+#
+# This file is part of ASRT Speech Recognition Tool.
+#
+# ASRT is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+# ASRT is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with ASRT.  If not, see <https://www.gnu.org/licenses/>.
+# ============================================================================
+
+"""
+@author: nl8590687
+ops.py单元测试
+"""
+
+import pytest
+from ops import get_edit_distance, ctc_decode_delete_tail_blank, ctc_decode_stream
+
+
+class TestGetEditDistance:
+    def test_1(self):
+        examle_input = ["abc", "abc"]
+        examle_output = 0
+        result = get_edit_distance(examle_input[0], examle_input[1])
+        assert result == examle_output
+
+    def test_2(self):
+        examle_input = ["abc", "adc"]
+        examle_output = 1
+        result = get_edit_distance(examle_input[0], examle_input[1])
+        assert result == examle_output
+
+    def test_3(self):
+        examle_input = ["abc", "a"]
+        examle_output = 2
+        result = get_edit_distance(examle_input[0], examle_input[1])
+        assert result == examle_output
+
+    def test_4(self):
+        examle_input = ["abc", "addce"]
+        examle_output = 3
+        result = get_edit_distance(examle_input[0], examle_input[1])
+        assert result == examle_output
+
+    def test_5(self):
+        examle_input = ["abc", ""]
+        examle_output = 3
+        result = get_edit_distance(examle_input[0], examle_input[1])
+        assert result == examle_output
+
+    def test_6(self):
+        examle_input = ["", ""]
+        examle_output = 0
+        result = get_edit_distance(examle_input[0], examle_input[1])
+        assert result == examle_output
+
+
+class TestCtcDecodeDeleteTailBlank:
+    def test_1(self):
+        examle_input = [1, 2, 3, 4, 5, -1, -1, -1, -1, -1, -1]
+        examle_output = [1, 2, 3, 4, 5]
+        result = ctc_decode_delete_tail_blank(examle_input)
+        assert result == examle_output
+
+    def test_2(self):
+        examle_input = [1, 2, 3, 4, 5, -1]
+        examle_output = [1, 2, 3, 4, 5]
+        result = ctc_decode_delete_tail_blank(examle_input)
+        assert result == examle_output
+
+    def test_3(self):
+        examle_input = [1, 2, 3, 4, 5]
+        examle_output = [1, 2, 3, 4, 5]
+        result = ctc_decode_delete_tail_blank(examle_input)
+        assert result == examle_output
+
+    def test_4(self):
+        examle_input = [-1, -1, -1, -1]
+        examle_output = []
+        result = ctc_decode_delete_tail_blank(examle_input)
+        assert result == examle_output
+
+    def test_5(self):
+        examle_input = [-1]
+        examle_output = []
+        result = ctc_decode_delete_tail_blank(examle_input)
+        assert result == examle_output
+
+
+class TestCtcDecodeStream:
+    def test_1(self):
+        examle_input = [-1, -1, -1, 1, 1, 1, -1, -1, 2, 2, 2, 3, 3, -1, -1, -1]
+        examle_output = [1, 2, 3, -1]
+        result = []
+        res, tmp = ctc_decode_stream(examle_input)
+        result.append(res)
+        while len(tmp) > 0:
+            res, tmp = ctc_decode_stream(tmp)
+            result.append(res)
+        assert result == examle_output
+
+    def test_2(self):
+        examle_input = [-1, -1, -1, -1, -1]
+        examle_output = [-1]
+        result = []
+        res, tmp = ctc_decode_stream(examle_input)
+        result.append(res)
+        while len(tmp) > 0:
+            res, tmp = ctc_decode_stream(tmp)
+            result.append(res)
+        assert result == examle_output

utils/thread.py

@@ -0,0 +1,55 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+#
+# Copyright 2016-2099 Ailemon.net
+#
+# This file is part of ASRT Speech Recognition Tool.
+#
+# ASRT is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+# ASRT is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with ASRT.  If not, see <https://www.gnu.org/licenses/>.
+# ============================================================================
+
+"""
+@author: nl8590687
+解决python生成器多线程的线程安全问题
+"""
+
+import threading
+'''
+    A generic iterator and generator that takes any iterator and wrap it to make it thread safe.
+    This method was introducted by Anand Chitipothu in http://anandology.com/blog/using-iterators-and-generators/
+    but was not compatible with python 3. This modified version is now compatible and works both in python 2.8 and 3.0 
+'''
+
+
+class threadsafe_iter:
+    """Takes an iterator/generator and makes it thread-safe by
+    serializing call to the `next` method of given iterator/generator.
+    """
+    def __init__(self, it):
+        self.it = it
+        self.lock = threading.Lock()
+
+    def __iter__(self):
+        return self
+
+    def __next__(self):
+        with self.lock:
+            return self.it.__next__()
+
+
+def threadsafe_generator(f):
+    """A decorator that takes a generator function and makes it thread-safe.
+    """
+    def g(*a, **kw):
+        return threadsafe_iter(f(*a, **kw))
+    return g