Fish Audio S2 - Real Expressive AI Voices
by•
We've open-sourced Fish Audio S2, a new generation of expressive TTS that lets you direct voices with natural language. Add cues like [whisper] or [laughing nervously], generate multi-speaker dialogue in one pass, and create scary-real voices across 80+ languages.

Replies
Congrats on the launch! I'm curious: if I’m building a real-time voice agent where latency and fine-grained emotion are dealbreakers, what specific benchmarks or features make Fish Audio a better bet than ElevenLabs right now?
Just came here to say I found Fish Audio via some deep research I had Opus 4.6 do on the best TTS providers and after several rounds of discussion about the findings, it was clear that Fish was the winner. Starting to integrate it into our AI Mental Training app now, and so far so good!
I must say that S2 is a game changer! For the small tasks it is running really well on a 5060 TI 16GB version and the bigger ones are able to be sourced to the cloud. Congratulations on you launch! And keep up the good work!
I had fun using this the first time i used and right now i will use it for some RP if my credits Refreshed, I love this as a voice acting and roleplay enthusiast, keep this going because i am looking forward for how much it improves
Es genial, he podido hacer muchos vídeos gracias a esta herramienta y junto a mi equipo le debemos mucho
動画編集に使用してます!ありがとうございます!★
I have access to ElevenLabs. But since I'm filming a feature film using neural networks, I need to voice emotional lines (f.e.«LISTEN! I don't give a CRAP about your ulcer!!! Or your sixty years!!! GET OUT! I CANT DO THIS ANYMORE!!! PACK YOUR STUFF!!! YOURE OUT OF HERE TODAY!!!»). After long attempts to voice my character's angry tirade (with extensive tone changes and the insertion of pauses and stresses in REAPER v.7), I began actively searching for another service. And I'm extremely glad I finally found Fish.audio!
Fish audio really surprised me it is a really good and accurate ai voice thingy. And id use it all the time. Iff.. there werent credits needed
Exelente trabajo, se abre una gran cantidad de usos y variedades sumando a qué tiene una exelente inclusión de sentimientos lo cual la hace más natural su resultado
¡Es increíble el como las voces son perfectamente interpretadas! Es como si estuviera hablando el personaje y no una IA