Uyghur Speech Models

Check out this demo to try the models from this repository.

Notebooks in this repo containing model fine-tuning code:

Fine_Tune_MMS_ASR_Uyghur: Meta-MMS-ASR fine-tuned for Uyghur with THUYG20, Common Voice, and UQSpeechDataset. The resulting model can be accessed here.
Fine_Tune_MMS_TTS_Uyghur: Meta-MMS-TTS fine-tuned for Uyghur with UQSpeechDataset. The resulting model can be accessed here.
Fine_Tune_Whisper_Uyghur_CV: OpenAI-Whisper fine-tuned for Uyghur with Common Voice. The resulting model can be accessed here.
Fine_Tune_Whisper_Uyghur_THUYG20: OpenAI-Whisper fine-tuned for Uyghur with THUYG20. The resulting model can be accessed here.

Notebooks contain examples of inferencing the ASR and TTS models:

Name		Name	Last commit message	Last commit date
Latest commit History 9 Commits
Fine_Tune_MMS_ASR_Uyghur.ipynb		Fine_Tune_MMS_ASR_Uyghur.ipynb
Fine_Tune_MMS_TTS_Uyghur.ipynb		Fine_Tune_MMS_TTS_Uyghur.ipynb
Fine_Tune_Whisper_Uyghur_CV.ipynb		Fine_Tune_Whisper_Uyghur_CV.ipynb
Fine_Tune_Whisper_Uyghur_THUYG20.ipynb		Fine_Tune_Whisper_Uyghur_THUYG20.ipynb
ReadMe.md		ReadMe.md
Uyghur_ASR_Inference.ipynb		Uyghur_ASR_Inference.ipynb
Uyghur_TTS_Inference.ipynb		Uyghur_TTS_Inference.ipynb

Provide feedback