search

WangHelin1997 / SSR-Speech

SSR-Speech: Towards Stable, Safe and Robust Zero-shot Speech Editing and Synthesis
https://wanghelin1997.github.io/SSR-Speech-Demo/
MIT License
72 stars 8 forks source link

Adjust output audio speed #8

Open jerrymatjila opened 3 days ago

jerrymatjila commented 3 days ago

Is there a way to adjust output audio speed when performing zero-shot TTS?

WangHelin1997 commented 3 days ago

Hi, Currently, the model cannot control the speed. However, different random seeds or hyperparameters could generate speech with different speed. You could adjust sub_amount, top_p, seed or cfg_coef in inference.py to get various results.