Yue Leng

Fish Speech - Few-shot voice cloning and text-to-speech

by
With just 15 seconds of any voice, Fish Speech can reliably synthesize natural and fluent speech while maintaining the given timbre, style, and accent. Our open-source team, creators of So-VITS-SVC and Bert-VITS2, proudly introduces Fish Speech.

Add a comment

Replies

Best
Yue Leng
We're excited to announce the launch of Fish Speech 1.3 on our Fish Audio website! We'll continue our commitment to open-sourcing TTS models. Join the discussion and check it out!