Update README.md
Browse files
README.md
CHANGED
|
@@ -4,6 +4,12 @@ language:
|
|
| 4 |
- de
|
| 5 |
base_model:
|
| 6 |
- HKUSTAudio/Llasa-1B-Multilingual
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7 |
---
|
| 8 |
<img src="https://huggingface.co/MultiLlasa/Llasa-1B-Multilingual-German/resolve/main/cover.webp" alt="Indic Parler Logo" width="800" style="margin-left:'auto' margin-right:'auto' display:'block'"/>
|
| 9 |
|
|
@@ -20,6 +26,8 @@ base_model:
|
|
| 20 |
|
| 21 |
This text-to-speech (TTS) model has been trained on a custom dataset representing **7,000 hours** of high-quality audio data. The audio data consisted of permissive podcasts, lectures and other OER data.
|
| 22 |
|
|
|
|
|
|
|
| 23 |
## Training Details
|
| 24 |
|
| 25 |
- **Base Model:** HKUSTAudio/Llasa-1B-Multilingual
|
|
|
|
| 4 |
- de
|
| 5 |
base_model:
|
| 6 |
- HKUSTAudio/Llasa-1B-Multilingual
|
| 7 |
+
widget:
|
| 8 |
+
- src: https://huggingface.co/MultiLlasa/Llasa-1B-Multilingual-German/resolve/main/examples/no_speaker_example.wav
|
| 9 |
+
example_title: Without reference audio
|
| 10 |
+
- src: https://huggingface.co/MultiLlasa/Llasa-1B-Multilingual-German/resolve/main/examples/with_reference.wav
|
| 11 |
+
example_title: With reference audio
|
| 12 |
+
|
| 13 |
---
|
| 14 |
<img src="https://huggingface.co/MultiLlasa/Llasa-1B-Multilingual-German/resolve/main/cover.webp" alt="Indic Parler Logo" width="800" style="margin-left:'auto' margin-right:'auto' display:'block'"/>
|
| 15 |
|
|
|
|
| 26 |
|
| 27 |
This text-to-speech (TTS) model has been trained on a custom dataset representing **7,000 hours** of high-quality audio data. The audio data consisted of permissive podcasts, lectures and other OER data.
|
| 28 |
|
| 29 |
+
|
| 30 |
+
|
| 31 |
## Training Details
|
| 32 |
|
| 33 |
- **Base Model:** HKUSTAudio/Llasa-1B-Multilingual
|