Spaces:

---

RinggAI

/

Ringg-TTS-v1.0

like 80

Running

Afterwards, there is an opportunity to release the base model, with less training data and shorter training time, but it still has some effect. Subsequent fine-tuning is possible.

Got it So you’re planning to release a version that supports fine-tuning with custom voices like shorter recordings and quicker training time right?

This is a zero-shot TTS model, so you can clone your voice with just 3 to 5 seconds of audio. However, if you're not satisfied with the voice rhythm, you can fine-tune it with more data.

Hi @samxiao0 we are working on some models internally as well and would release the model with voice cloning capabilities sometime in future