Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
@@ -182,25 +182,25 @@ description = """
|
|
182 |
The <b>SpeechT5</b> model is pre-trained on text as well as speech inputs, with targets that are also a mix of text and speech.
|
183 |
By pre-training on text and speech at the same time, it learns unified representations for both, resulting in improved modeling capabilities.
|
184 |
|
185 |
-
SpeechT5 can be fine-tuned for different speech tasks. This space demonstrates the <b>text-to-speech</b> (TTS) checkpoint for the
|
186 |
|
187 |
See also the <a href="https://huggingface.co/spaces/Matthijs/speecht5-asr-demo">speech recognition (ASR) demo</a>
|
188 |
and the <a href="https://huggingface.co/spaces/Matthijs/speecht5-vc-demo">voice conversion demo</a>.
|
189 |
|
190 |
Refer to <a href="https://colab.research.google.com/drive/1i7I5pzBcU3WDFarDnzweIj4-sVVoIUFJ">this Colab notebook</a> to learn how to fine-tune the SpeechT5 TTS model on your own dataset or language.
|
191 |
|
192 |
-
<b>How to use:</b> Enter some
|
193 |
HiFi-GAN vocoder. Because the model always applies random dropout, each attempt will give slightly different results.
|
194 |
The <em>Surprise Me!</em> option creates a completely randomized speaker.
|
195 |
"""
|
196 |
|
197 |
examples = [
|
198 |
-
["
|
199 |
-
["
|
200 |
-
["
|
201 |
-
["
|
202 |
-
["
|
203 |
-
["
|
204 |
]
|
205 |
|
206 |
gr.Interface(
|
|
|
182 |
The <b>SpeechT5</b> model is pre-trained on text as well as speech inputs, with targets that are also a mix of text and speech.
|
183 |
By pre-training on text and speech at the same time, it learns unified representations for both, resulting in improved modeling capabilities.
|
184 |
|
185 |
+
SpeechT5 can be fine-tuned for different speech tasks. This space demonstrates the <b>text-to-speech</b> (TTS) checkpoint for the Armenian language.
|
186 |
|
187 |
See also the <a href="https://huggingface.co/spaces/Matthijs/speecht5-asr-demo">speech recognition (ASR) demo</a>
|
188 |
and the <a href="https://huggingface.co/spaces/Matthijs/speecht5-vc-demo">voice conversion demo</a>.
|
189 |
|
190 |
Refer to <a href="https://colab.research.google.com/drive/1i7I5pzBcU3WDFarDnzweIj4-sVVoIUFJ">this Colab notebook</a> to learn how to fine-tune the SpeechT5 TTS model on your own dataset or language.
|
191 |
|
192 |
+
<b>How to use:</b> Enter some Armenian text and choose a speaker. The output is a mel spectrogram, which is converted to a mono 16 kHz waveform by the
|
193 |
HiFi-GAN vocoder. Because the model always applies random dropout, each attempt will give slightly different results.
|
194 |
The <em>Surprise Me!</em> option creates a completely randomized speaker.
|
195 |
"""
|
196 |
|
197 |
examples = [
|
198 |
+
["Մեր ճակատագիրը աստղերի մեջ չէ, այլ մեր մեջ:", "BDL (male)"],
|
199 |
+
["Հոկտեմբերին ութոտնուկն ու Օլիվերը գնացին օպերա։", "BDL (male)],
|
200 |
+
["Նա ծովի ափին ծովախեցգետիններ է վաճառում: Ես տեսա, որ խոհանոցում հավ է ուտում մի ձագ:", "BDL (male)],
|
201 |
+
["Կտրուկ խիզախ բրիգադները թափահարում էին լայն, պայծառ շեղբեր, կոպիտ ավտոբուսներ և մռութներ՝ վատ հավասարակշռելով դրանք:", "BDL (male)],
|
202 |
+
["Դարչինի հոմանիշը դարչինի հոմանիշն է:", "BDL (male)"],
|
203 |
+
["Ինչքա՞ն փայտ կթափի փայտափայտը, եթե փայտափայտը կարողանար փայտ ծակել: Նա կխփեր, կաներ, այնքան, որքան կարող էր, և այնքան փայտ կխփեր, որքան փայտափայտը, եթե փայտափայտը կարողանար փայտ ծակել:", "BDL (male)],
|
204 |
]
|
205 |
|
206 |
gr.Interface(
|