urdu ai tts
Dear OpenMOSS Team / Owners,
First of all, huge congratulations on the amazing release of MOSS-TTS-Nano. The fact that this model runs so smoothly on a CPU with zero-shot voice cloning capabilities is truly groundbreaking!
I am writing to make a very special and earnest request: Please prioritize adding support for the Urdu and Hindi languages in your upcoming updates. I live in Pakistan, and Urdu is my native language. To be completely honest, while there are a few Urdu TTS models out there, they either sound incredibly robotic, or the good ones require massive, expensive GPUs to run.
The true magic of your MOSS-TTS-Nano is its flawless CPU performance. That is exactly what our community desperately needs! I highly request you to build a mixed model that seamlessly supports English, Urdu, and Hindi together.
I am currently developing my own AI Web UI under my company name, NZG (nzg73), and it is my absolute dream to integrate your CPU-friendly model into my release.
To save your team some time, I have done the research and found exactly what you need. Here are the best, high-quality open-source datasets available right now for these languages:
For Urdu:
https://huggingface.co/datasets/humairawan/Urdu-NSW/tree/main
https://huggingface.co/datasets/humairawan/UrduSpeech/tree/main/Urdu
https://huggingface.co/datasets/humair025/Urdu-ONYX-WAV/tree/main
https://huggingface.co/datasets/ai4bharat/indicvoices_r/tree/main/Urdu
For Hindi:
https://huggingface.co/datasets/ai4bharat/indicvoices_r/tree/main/Hindi
I will be honest: I am not a highly technical genius or a dataset expert. However, because Urdu is my native language and I know its exact natural flow, I can offer my personal help. If you need clean, native audio data for training or testing, I am more than happy to record my own voice and provide you with high-quality, studio-level audio samples. Just let me know what you need!
Lastly, I have one more dream request for your roadmap: Please consider releasing a unified CPU-friendly model in the future that handles both Text-to-Speech (TTS) and Speech-to-Text (STT) simultaneously. A unified, lightweight CPU model like this doesn't exist yet!
Please make the Urdu model powerful. This is my sincere request to you. Thank you for your incredible contribution to the open-source AI community!
Best regards,
NZG (nzg73) Email: nzgnzg73@gmail.com