is it possible to integrate with Supertonic lightweight TTS?

[Willy030125]

This local offline TTS is amazing support, it would be amazing if it can be integrated with Rkllama. And it supports 31 languages:
https://github.com/supertone-inc/supertonic

could be an alternative choice for TTS besides Piper.

[Willy030125]

https://supertone-inc.github.io/supertonic-py/cli/serve/
https://supertone-inc.github.io/supertonic-py/voices/

[danielferr85]

I think its is possible. It is only needed to create a .py file (like supertonic.py) in the package rkllama/api/models/audio with the logic and include this model in the IF section in the file rkllama/api/tts.py.

An user already converted this model in HF and include the test code for the model https://huggingface.co/Sail2Dream/supertonic-rknn-rk3588 so it is possible.

I'm currently very busy but maybe someone can implement this... Or maybe ask for any LLM to do it with opencode, copilot, etc.... =)