Update README.md
Browse files
README.md
CHANGED
@@ -8,7 +8,7 @@ base_model:
|
|
8 |
pipeline_tag: text-to-speech
|
9 |
---
|
10 |
|
11 |
-
#
|
12 |

|
13 |
|
14 |
## Table of Contents
|
@@ -25,7 +25,7 @@ pipeline_tag: text-to-speech
|
|
25 |
|
26 |
## Model Summary
|
27 |
|
28 |
-
|
29 |
|
30 |
<video controls width="600">
|
31 |
<source src="https://huggingface.co/saheedniyi/YarnGPT/resolve/main/audio/YearnGPT.mp4" type="video/mp4">
|
@@ -33,20 +33,15 @@ YarnGPT is a text-to-speech (TTS) model designed to synthesize Nigerian-accented
|
|
33 |
</video>
|
34 |
|
35 |
#### How to use (Colab)
|
36 |
-
The model can generate audio on its own but its better to use a voice to prompt the model
|
37 |
-
|
38 |
-
|
39 |
-
- tayo
|
40 |
-
-
|
41 |
-
-
|
42 |
-
-
|
43 |
-
|
44 |
-
|
45 |
-
- osagie
|
46 |
-
- joke
|
47 |
-
- emma (the names do not correlate to any tribe or accent)
|
48 |
-
|
49 |
-
### Prompt YarnGPT
|
50 |
```python
|
51 |
# clone the YarnGPT repo to get access to the `audiotokenizer`
|
52 |
!git clone https://github.com/saheedniyi02/yarngpt.git
|
@@ -265,54 +260,84 @@ Listen to samples generated by YarnGPT:
|
|
265 |
</thead>
|
266 |
<tbody>
|
267 |
<tr>
|
268 |
-
<td style="border: 1px solid #ddd; padding: 8px;">
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
269 |
<td style="border: 1px solid #ddd; padding: 8px;">
|
270 |
<audio controls style="width: 100%;">
|
271 |
-
<source src="https://huggingface.co/saheedniyi/
|
272 |
Your browser does not support the audio element.
|
273 |
</audio>
|
274 |
</td>
|
275 |
-
<td style="border: 1px solid #ddd; padding: 8px;">(temperature=0.1, repetition_penalty=1.1), voice:
|
276 |
</tr>
|
277 |
<tr>
|
278 |
-
<td style="border: 1px solid #ddd; padding: 8px;"
|
279 |
<td style="border: 1px solid #ddd; padding: 8px;">
|
280 |
<audio controls style="width: 100%;">
|
281 |
-
<source src="https://huggingface.co/saheedniyi/
|
282 |
Your browser does not support the audio element.
|
283 |
</audio>
|
284 |
</td>
|
285 |
-
<td style="border: 1px solid #ddd; padding: 8px;">(temperature=0.1, repetition_penalty=1.1), voice:
|
286 |
</tr>
|
287 |
<tr>
|
288 |
-
<td style="border: 1px solid #ddd; padding: 8px;">
|
289 |
<td style="border: 1px solid #ddd; padding: 8px;">
|
290 |
<audio controls style="width: 100%;">
|
291 |
-
<source src="https://huggingface.co/saheedniyi/
|
292 |
Your browser does not support the audio element.
|
293 |
</audio>
|
294 |
</td>
|
295 |
-
<td style="border: 1px solid #ddd; padding: 8px;">(temperature=0.1, repetition_penalty=1.1),
|
296 |
</tr>
|
297 |
<tr>
|
298 |
-
<td style="border: 1px solid #ddd; padding: 8px;">
|
299 |
<td style="border: 1px solid #ddd; padding: 8px;">
|
300 |
<audio controls style="width: 100%;">
|
301 |
-
<source src="https://huggingface.co/saheedniyi/
|
302 |
Your browser does not support the audio element.
|
303 |
</audio>
|
304 |
</td>
|
305 |
-
<td style="border: 1px solid #ddd; padding: 8px;">(temperature=0.1, repetition_penalty=1.1), voice:
|
306 |
</tr>
|
307 |
<tr>
|
308 |
-
<td style="border: 1px solid #ddd; padding: 8px;">
|
309 |
<td style="border: 1px solid #ddd; padding: 8px;">
|
310 |
<audio controls style="width: 100%;">
|
311 |
-
<source src="https://huggingface.co/saheedniyi/
|
312 |
Your browser does not support the audio element.
|
313 |
</audio>
|
314 |
</td>
|
315 |
-
<td style="border: 1px solid #ddd; padding: 8px;">(temperature=0.1, repetition_penalty=1.1)</td>
|
316 |
</tr>
|
317 |
</tbody>
|
318 |
</table>
|
|
|
8 |
pipeline_tag: text-to-speech
|
9 |
---
|
10 |
|
11 |
+
# YarnGPT2
|
12 |

|
13 |
|
14 |
## Table of Contents
|
|
|
25 |
|
26 |
## Model Summary
|
27 |
|
28 |
+
YarnGPT2 is a text-to-speech (TTS) model designed to synthesize Nigerian-accented Languages (yoruba, igbo, hausa and english) leveraging pure language modelling without external adapters or complex architectures, offering high-quality, natural, and culturally relevant speech synthesis for diverse applications.
|
29 |
|
30 |
<video controls width="600">
|
31 |
<source src="https://huggingface.co/saheedniyi/YarnGPT/resolve/main/audio/YearnGPT.mp4" type="video/mp4">
|
|
|
33 |
</video>
|
34 |
|
35 |
#### How to use (Colab)
|
36 |
+
The model can generate audio on its own but its better to use a voice to prompt the model:
|
37 |
+
|
38 |
+
##### Voices (arranged in order of perfomance and stability)
|
39 |
+
- English: idera, chinenye, jude, emma,umar,,joke,zainab ,osagie, remi, tayo
|
40 |
+
- Yoruba: yoruba_male2, yoruba_female2, yoruba_feamle1
|
41 |
+
- Igbo: igbo_female2, igbo_male2,igbo_female1,
|
42 |
+
- Hausa: hausa_feamle1,hausa_female2, hausa_male2,hausa_male1
|
43 |
+
|
44 |
+
### Prompt YarnGPT2
|
|
|
|
|
|
|
|
|
|
|
45 |
```python
|
46 |
# clone the YarnGPT repo to get access to the `audiotokenizer`
|
47 |
!git clone https://github.com/saheedniyi02/yarngpt.git
|
|
|
260 |
</thead>
|
261 |
<tbody>
|
262 |
<tr>
|
263 |
+
<td style="border: 1px solid #ddd; padding: 8px;">Uhm, so, what was the inspiration behind your latest project? Like, was there a specific moment where you were like, 'Yeah, this is it!' Or, you know, did it just kind of, uh, come together naturally over time</td>
|
264 |
+
<td style="border: 1px solid #ddd; padding: 8px;">
|
265 |
+
<audio controls style="width: 100%;">
|
266 |
+
<source src="https://huggingface.co/saheedniyi/YarnGPT2/resolve/main/Audio/Audio1.wav" type="audio/wav">
|
267 |
+
Your browser does not support the audio element.
|
268 |
+
</audio>
|
269 |
+
</td>
|
270 |
+
<td style="border: 1px solid #ddd; padding: 8px;">(temperature=0.1, repetition_penalty=1.1), language: english, voice: idera</td>
|
271 |
+
</tr>
|
272 |
+
<tr>
|
273 |
+
<td style="border: 1px solid #ddd; padding: 8px;">The election was won by businessman and politician, Moshood Abiola, but Babangida annulled the results, citing concerns over national security.</td>
|
274 |
+
<td style="border: 1px solid #ddd; padding: 8px;">
|
275 |
+
<audio controls style="width: 100%;">
|
276 |
+
<source src="https://huggingface.co/saheedniyi/YarnGPT2/resolve/main/Audio/Audio2.wav" type="audio/wav">
|
277 |
+
Your browser does not support the audio element.
|
278 |
+
</audio>
|
279 |
+
</td>
|
280 |
+
<td style="border: 1px solid #ddd; padding: 8px;">(temperature=0.1, repetition_penalty=1.1), language: english, voice: zainab</td>
|
281 |
+
</tr>
|
282 |
+
<tr>
|
283 |
+
<td style="border: 1px solid #ddd; padding: 8px;">Habeeb Okikiọla Olalomi Badmus ti ọpọ awọn ololufẹ rẹ mọ si Portable ti sọ fun ile ẹjọ majisireeti ti ipinlẹ Ogun wi pe ṣaka lara oun da, oun ko ni aisan tabi arun kankan lara.</td>
|
284 |
+
<td style="border: 1px solid #ddd; padding: 8px;">
|
285 |
+
<audio controls style="width: 100%;">
|
286 |
+
<source src="https://huggingface.co/saheedniyi/YarnGPT2/resolve/main/Audio/Audio3.wav" type="audio/wav">
|
287 |
+
Your browser does not support the audio element.
|
288 |
+
</audio>
|
289 |
+
</td>
|
290 |
+
<td style="border: 1px solid #ddd; padding: 8px;">(temperature=0.1, repetition_penalty=1.1), language: yoruba, voice: yoruba_male2</td>
|
291 |
+
</tr>
|
292 |
+
<tr>
|
293 |
+
<td style="border: 1px solid #ddd; padding: 8px;">Gómìnà náà fẹ̀sùn kàn pé àwọn alága àná gbìyànjú láti fi ipá gba àwọn ìjọba ìbílẹ̀ lọ́nà àìtọ́, tó sì jẹ́ pé ó yẹ kí àwọn ìjọba ìbílẹ̀ náà wà ní títì</td>
|
294 |
<td style="border: 1px solid #ddd; padding: 8px;">
|
295 |
<audio controls style="width: 100%;">
|
296 |
+
<source src="https://huggingface.co/saheedniyi/YarnGPT2/resolve/main/Audio/Audio4.wav" type="audio/wav">
|
297 |
Your browser does not support the audio element.
|
298 |
</audio>
|
299 |
</td>
|
300 |
+
<td style="border: 1px solid #ddd; padding: 8px;">(temperature=0.1, repetition_penalty=1.1), language: yoruba, voice: yoruba_female2</td>
|
301 |
</tr>
|
302 |
<tr>
|
303 |
+
<td style="border: 1px solid #ddd; padding: 8px;">Ọ bụ oge ha si Enugwu steeti eme njem aga Anambra ka ndị omekome ahụ wakporo ụgbọala ha.</td>
|
304 |
<td style="border: 1px solid #ddd; padding: 8px;">
|
305 |
<audio controls style="width: 100%;">
|
306 |
+
<source src="https://huggingface.co/saheedniyi/YarnGPT2/resolve/main/Audio/Audio5.wav" type="audio/wav">
|
307 |
Your browser does not support the audio element.
|
308 |
</audio>
|
309 |
</td>
|
310 |
+
<td style="border: 1px solid #ddd; padding: 8px;">(temperature=0.1, repetition_penalty=1.1), language: igbo, voice: igbo_male2</td>
|
311 |
</tr>
|
312 |
<tr>
|
313 |
+
<td style="border: 1px solid #ddd; padding: 8px;">Isi ụlọorụ Shell dị na Lọndọn na gọọmenti Naịjirịa ekwuputala ugboro ugboro na ọrụ ịsacha ogbe ndị lara n'iyi n'Ogoni bụ nke malitere ihe dịka afọ asatọ gara aga na-aga nke ọma.</td>
|
314 |
<td style="border: 1px solid #ddd; padding: 8px;">
|
315 |
<audio controls style="width: 100%;">
|
316 |
+
<source src="https://huggingface.co/saheedniyi/YarnGPT2/resolve/main/Audio/Audio6.wav" type="audio/wav">
|
317 |
Your browser does not support the audio element.
|
318 |
</audio>
|
319 |
</td>
|
320 |
+
<td style="border: 1px solid #ddd; padding: 8px;">(temperature=0.1, repetition_penalty=1.1), language: igbo, voice: igbo_female1</td>
|
321 |
</tr>
|
322 |
<tr>
|
323 |
+
<td style="border: 1px solid #ddd; padding: 8px;">Gwamnatin Najeriya ta sake maka shafin hada-hadar kuɗin kirifto na Binance a kotu, inda take buƙatar ya biya ta diyyar kuɗi dalar Amurka biliyan 81.5</td>
|
324 |
<td style="border: 1px solid #ddd; padding: 8px;">
|
325 |
<audio controls style="width: 100%;">
|
326 |
+
<source src="https://huggingface.co/saheedniyi/YarnGPT2/resolve/main/Audio/Audio7.wav" type="audio/wav">
|
327 |
Your browser does not support the audio element.
|
328 |
</audio>
|
329 |
</td>
|
330 |
+
<td style="border: 1px solid #ddd; padding: 8px;">(temperature=0.1, repetition_penalty=1.1), language: hausa, voice: hausa_female1</td>
|
331 |
</tr>
|
332 |
<tr>
|
333 |
+
<td style="border: 1px solid #ddd; padding: 8px;">Bisa ga dukkan alamu, haƙata cimma ruwa, dangane da koke-koken da tsofaffin ma'aikatan tarayya ke ta yi, a kan dimbin basukan wasu hakkokinsu da suke bi shekara da shekaru.</td>
|
334 |
<td style="border: 1px solid #ddd; padding: 8px;">
|
335 |
<audio controls style="width: 100%;">
|
336 |
+
<source src="https://huggingface.co/saheedniyi/YarnGPT2/resolve/main/Audio/Audio8.wav" type="audio/wav">
|
337 |
Your browser does not support the audio element.
|
338 |
</audio>
|
339 |
</td>
|
340 |
+
<td style="border: 1px solid #ddd; padding: 8px;">(temperature=0.1, repetition_penalty=1.1), language: hausa, voice: hausa_male2</td>
|
341 |
</tr>
|
342 |
</tbody>
|
343 |
</table>
|