Hugging Face's logo

Datasets:
ganga4364
/
STT_training_with_metadata

License:
Dataset card Data Studio Files Community
1

Update_Data

#1
by ganga4364 - opened
base: refs/heads/main
←
from: refs/pr/1
Discussion Files changed
+0
-226
Files changed (9) hide show
  1. .gitattributes +0 -8
  2. AB_train.csv +0 -3
  3. CS_train.csv +0 -3
  4. HS_train.csv +0 -3
  5. MV_train.csv +0 -3
  6. NS_train.csv +0 -3
  7. NW_train.csv +0 -3
  8. PC_train.csv +0 -3
  9. README.md +0 -197
.gitattributes CHANGED
@@ -56,11 +56,3 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
56
  # Video files - compressed
57
  *.mp4 filter=lfs diff=lfs merge=lfs -text
58
  *.webm filter=lfs diff=lfs merge=lfs -text
59
- AB_train.csv filter=lfs diff=lfs merge=lfs -text
60
- CS_train.csv filter=lfs diff=lfs merge=lfs -text
61
- HS_train.csv filter=lfs diff=lfs merge=lfs -text
62
- MV_train.csv filter=lfs diff=lfs merge=lfs -text
63
- NS_train.csv filter=lfs diff=lfs merge=lfs -text
64
- NW_train.csv filter=lfs diff=lfs merge=lfs -text
65
- PC_train.csv filter=lfs diff=lfs merge=lfs -text
66
- TT_train.csv filter=lfs diff=lfs merge=lfs -text
 
56
  # Video files - compressed
57
  *.mp4 filter=lfs diff=lfs merge=lfs -text
58
  *.webm filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
 
AB_train.csv DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:87071c76a0b250eb868f78bbcef9089a0dcd2daa1b77838b9b9d0d87fc7c6639
3
- size 105397698
 
 
 
 
CS_train.csv DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:00525a7a25ab8769e5086a29b11f64ab199b34c0c2c18c4ecc18bda4cdd591eb
3
- size 89970565
 
 
 
 
HS_train.csv DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:c9b04da0c7ab4020ec48c9b6b7419e39fadcbc99ccbb5ae8ce3e8b4532a9c3de
3
- size 43075807
 
 
 
 
MV_train.csv DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:1c28b284da8479de990035f31926111dfc696e84d21cd1e4317611057a2ea52b
3
- size 64337256
 
 
 
 
NS_train.csv DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:af7665a602db8f1802a90b9f78b565f5349c6d27f4dbe299483d3c48aa2cbe2e
3
- size 226072840
 
 
 
 
NW_train.csv DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:d2e916645ea87384073f8ba20b2d118df0a33156387c01522f7a56417347d847
3
- size 68395897
 
 
 
 
PC_train.csv DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:ccb88cc45dea43a0612f6d25416473a65bec41b593e92986c58412fdc61c9547
3
- size 51070253
 
 
 
 
README.md DELETED
@@ -1,197 +0,0 @@
1
- ---
2
- license: mit
3
- ---
4
-
5
- # STT Training Data
6
-
7
- This dataset is prepared for Speech-to-Text (STT) training. It contains audio files from different departments, each annotated with their transcription. The dataset is split by department and includes the duration of audio data per department.
8
-
9
- ## Dataset Overview
10
-
11
- - **Total Audio Files**: 1,391,020
12
- - **Total Audio Length**: 1,576.76 hours
13
- - **Total Audio collected data**: 18 OCT, 2024
14
-
15
- ## Department-wise Statistics
16
-
17
- | Department | Number of Audio Files | Total Audio Length (hours) |
18
- |------------|-----------------------|----------------------------|
19
- | STT_AB | 174,525 | 224.92 |
20
- | STT_CS | 150,310 | 219.51 |
21
- | STT_HS | 87,088 | 96.27 |
22
- | STT_MV | 112,756 | 193.27 |
23
- | STT_NS | 241,007 | 318.28 |
24
- | STT_NW | 97,145 | 137.82 |
25
- | STT_PC | 73,894 | 78.90 |
26
- | STT_TT | 454,295 | 307.78 |
27
-
28
- ## Total Audio Length by Department
29
-
30
- The total audio length of each department in hours:
31
-
32
- - **STT_AB**: 224.92 hours
33
- - **STT_CS**: 219.51 hours
34
- - **STT_HS**: 96.27 hours
35
- - **STT_MV**: 193.27 hours
36
- - **STT_NS**: 318.28 hours
37
- - **STT_NW**: 137.82 hours
38
- - **STT_PC**: 78.90 hours
39
- - **STT_TT**: 307.78 hours
40
-
41
- ## Total Audio Length
42
-
43
- The combined audio length across all departments is **1,576.76 hours**.
44
-
45
-
46
- **Transcription Quality Grades:**
47
-
48
- | Grade | Description |
49
- |-------|---------------------------------|
50
- | 1 | Transcribed only |
51
- | 2 | Reviewed by the team lead |
52
- | 3 | Reviewed by the quality control team |
53
-
54
- ## Departments and Metadata
55
- Here is the structured table format for the metadata descriptions:
56
- ## STT_Training Metadata
57
-
58
- | Column | Description |
59
- |-------------|---------------------------------------------------------------------------------------------------|
60
- | **file_name** | Structured name of the audio file, including department, project ID, segment ID, and time range.|
61
- | **uni** | Tibetan transcript in Unicode. |
62
- | **wylie** | Wylie transliteration of the Tibetan text. |
63
- | **url** | URL link to the corresponding audio file. |
64
- | **dept** | Department identifier (e.g., STT_NS). |
65
- | **grade** | Quality grade of the transcription. |
66
- | **char_len** | Character length of the transcript. |
67
- | **audio_len** | Duration of the audio file in seconds. |
68
-
69
- ## STT_NS (Natural Speech) Metadata
70
-
71
- | Column | Description |
72
- |-------------------|-----------------------------------------------------------------------------------------------|
73
- | **Original ID** | Unique identifier of the source, if available. |
74
- | **Speaker ID** | Identifier for the speaker; multiple speakers are listed as a group. |
75
- | **Mic/Phone** | Type of recording device used (e.g., microphone, phone). |
76
- | **Location** | Geographical location of the speaker. |
77
- | **Topic** | Subject matter of the audio. |
78
- | **A Short Scenario** | Brief summary of the recording. |
79
- | **Speaker Name** | Name of the speaker. |
80
- | **Email ID** | Email address of the speaker. |
81
- | **Gender** | Gender of the speaker. |
82
- | **Home Address** | Speaker's home address. |
83
- | **Contact No** | Contact number of the speaker. |
84
- | **Age** | Age of the speaker. |
85
- | **Course** | Associated course, if relevant. |
86
- | **Year** | Year of the recording. |
87
- | **Nationality** | Nationality of the speaker. |
88
- | **Birth Place** | Speaker's birthplace. |
89
- | **Date of Birth** | Date of birth of the speaker. |
90
- | **Native Language** | First language of the speaker. |
91
- | **College** | Associated institution of the speaker. |
92
-
93
- ## STT_AB (Audio Book) Metadata
94
-
95
- | Column | Description |
96
- |----------------------|-------------------------------------------------------------------------------------------------|
97
- | **Original ID** | Unique identifier of the source, if available. |
98
- | **Collection** | Collection or set to which the audio belongs. |
99
- | **Book Title** | Title of the recorded book or text. |
100
- | **Category** | Classification of the recording (e.g., text). |
101
- | **Recorder** | Identifier of the person who recorded the audio. |
102
- | **Music** | Indicates if background music is present (True/False). |
103
- | **Name** | Speaker's name. |
104
- | **Gender** | Gender of the speaker. |
105
- | **Age** | Age of the speaker. |
106
- | **Place** | Recording location. |
107
- | **Livelihood** | Speaker's profession or occupation. |
108
- | **Birth Place** | Birthplace of the speaker. |
109
- | **Social/Marital Status** | Speaker’s social or marital status. |
110
- | **Education Level** | Educational background of the speaker. |
111
-
112
- ## STT_NW (News) Metadata
113
-
114
- | Column | Description |
115
- |----------------------|-------------------------------------------------------------------------------------------------|
116
- | **Name of News** | Title of the news segment. |
117
- | **Publishing Year** | Year the news was published. |
118
- | **Media Organization** | Media outlet responsible (e.g., TibetTV, VOA Tibetan). |
119
- | **Audio Quality** | Quality rating of the audio. |
120
- | **Single Speaker** | Indicates if a single speaker is present (True/False). |
121
- | **Country** | Country of the news production. |
122
- | **Original ID** | Unique identifier of the original source, if available. |
123
-
124
- ## STT_HS (Historical Speech) Metadata
125
-
126
- | Column | Description |
127
- |-------------------|-----------------------------------------------------------------------------------------------|
128
- | **file_name** | Structured name of the audio file, including department, project ID, segment ID, and time range. |
129
- | **uni** | Tibetan transcript in Unicode format, preserving the original speech content. |
130
- | **wylie** | Wylie transliteration of the Tibetan transcript, providing a phonetic representation. |
131
- | **url** | Direct URL link to the corresponding audio file stored on a cloud server. |
132
- | **dept** | Department identifier, specifically for Historical Speech recordings (STT_HS). |
133
- | **grade** | Quality grade of the transcription, indicating the level of review and quality assurance. |
134
- | **char_len** | The character length of the transcript, representing the size of the text content. |
135
- | **audio_len** | Duration of the audio file in seconds, indicating the length of the speech segment. |
136
- | **Publishing Year** | Year the speech was recorded or published, providing historical context to the content. |
137
- | **Name** | Name of the speaker, often with an alias to protect privacy. |
138
- | **Gender** | Gender of the speaker, typically categorized as Male or Female. |
139
- | **Age Group** | Age range of the speaker at the time of recording, categorized into groups. |
140
- | **Place of Origin** | Region or cultural area the speaker originates from, such as Kham or Utsang. |
141
- | **Exiled Year** | Year when the speaker went into exile, providing context to their personal history. |
142
- | **Original ID** | Unique identifier for the source or original collection of the recording. |
143
-
144
-
145
- ## STT_CS Children Speech Metadata
146
-
147
- | Column | Description |
148
- |---------------------|----------------------------------------------------------------------------------------------|
149
- | **Microphone_Type** | Type of microphone used for recording, if available. |
150
- | **Recording_Type** | The type of recording (e.g., classroom, outdoor, etc.). |
151
- | **Speakers ID** | Unique identifier for the child speaker. |
152
- | **Birth Year** | Year of birth of the child speaker, providing demographic context. |
153
- | **Gender** | Gender of the child speaker, typically categorized as Male or Female. |
154
- | **CITY** | City of residence of the child speaker, indicating their geographical background. |
155
- | **TOWN** | Town of residence, if applicable, providing more specific location details. |
156
- | **school** | Identifier for the school attended by the child, if available. |
157
- | **school name** | Name of the school or educational institution attended by the child. |
158
- | **Birth Place** | Birthplace of the child speaker, indicating their origin. |
159
- | **spoken tongue** | Primary language spoken by the child, providing context for their linguistic background. |
160
- | **Original ID** | Unique identifier for the source or original collection of the recording. |
161
- | **CLASS** | Academic class or grade level of the child at the time of the recording. |
162
-
163
-
164
- ## STT_MV Movie Speech Metadata
165
-
166
- | Column | Description |
167
- |---------------------|----------------------------------------------------------------------------------------------|
168
- | **Name of the movie** | Title of the movie from which the dialogue is extracted. |
169
- | **Subtitle** | Subtitle information for the dialogue, if available (e.g., "Chinese"). |
170
- | **Original ID** | Unique identifier for the source or original collection of the recording. |
171
-
172
-
173
- ## STT_PC Podcast Speech Metadata
174
-
175
- | Column | Description |
176
- |---------------------|-----------------------------------------------------------------------------------------------|
177
- | **Name of podcast** | Title of the podcast episode, highlighting the main topic discussed. |
178
- | **Publishing year** | Year the podcast episode was published, providing context to the content. |
179
- | **Channel** | Name of the channel or platform where the podcast is broadcasted (e.g., "Tibetan Libr"). |
180
- | **Audio Quality** | Rating of the audio quality, indicating clarity and production standards (e.g., "Good"). |
181
- | **Country** | Country of origin or production for the podcast episode (e.g., "India"). |
182
- | **Gender** | Gender of the speaker in the podcast episode (e.g., "Male"). |
183
- | **Age Group** | Age range of the speaker at the time of recording (e.g., "60-70"). |
184
- | **Original ID** | Unique identifier for the source or original collection of the recording. |
185
-
186
-
187
- ## STT_TT Tibetan Teaching Talk Metadata
188
-
189
- | Column | Description |
190
- |---------------------|-----------------------------------------------------------------------------------------------|
191
- | **Teacher** | Name of the teacher delivering the talk, often with an alias to protect privacy. |
192
- | **Topic** | Subject matter discussed in the talk, providing context to the content. |
193
- | **books** | Associated books or materials referenced in the talk, if applicable. |
194
- | **Episode** | Episode number or identifier for the talk series. |
195
- | **Original ID** | Unique identifier for the source or original collection of the recording. |
196
-
197
- ---