Request Multilingual TTS with Kokoro 1.0

[thuongvovan]

Hi team,

This project is truly amazing! The Kokoro 1.0 model is a multilingual TTS model. I propose we integrate it with MLX-Audio to build a versatile, natural-sounding multilingual speech system. Like here.

It has huge potential for multilingual content creation and accessibility. Looking forward to your thoughts!

One note: the current implementation seems to have an issue with the Chinese (zh) voice — it sounds closer to Cantonese rather than Mandarin.

Thank you

[thuongvovan]

Japanese seems to have a lot of wrong pronunciation

[Blaizzy]

Thank you very much!

We need to investigate this further but it's hard because I only speak Portuguese, Spanish, English, and a bit of Polish and Hindi.

Could you help me with example samples where it fails alongside the correct ones? That way it will be easier for me to debug.