Plain-AyI
← All projects
Speech & audio

Fish Speech

Open text-to-speech models that create expressive voices, clone voices and voice whole scenes from a script.

Research-backed

What it does

Fish Speech turns text into spoken audio. Its newest model, Drama 3, can voice several characters in one scene, clone a voice from a sample, and follow plain-language directions such as "gentler, and a bit sadder".

You can edit generated audio the way you edit text, for example swapping a single word while keeping the voice natural. Its makers say it handles long recordings and strong emotions like laughter, crying and whispers.

It can be run from the command line, as a server or through a web interface. A hosted version with a playground is available on the Fish Audio website.

Who it's for

Creators, game and audio producers, and developers who need natural, controllable AI voices.

Can I use it without coding?

With some setup. Needs installing and downloading model files. There's a web interface, and a powerful graphics card is recommended. New to this? Start here.

Can I run it?

Cost
FreeNothing to pay or sign up for.
Computer
Works best on a powerful computer or one with a graphics card (GPU)
Privacy
Runs on your own device: your data stays with you
API key
Not needed

Try it

  1. Follow the installation guide on the project's website to set it up and download the model files.
  2. Create speech using the command line, the web interface or the API server.
Official download page

Using Claude Code, Cursor or another AI coding assistant? Paste the prompt and it will walk you through the setup.New to this?

Before you install

  • Only download from the official GitHub page or the official download page above. Copies with similar names exist for many popular projects.
  • No standard license listed. Check the GitHub page before using it for business.
  • More on staying safe

Where people are talking about it