Plain-AyI
← All projects
Speech & audio

IndexTTS

Reads text aloud in a voice copied from one short audio clip, with control over emotion and speed.

Research-backed

What it does

IndexTTS is a text-to-speech system: it turns written text into spoken audio. Give it one short recording of a voice and it can speak new text in a similar-sounding voice.

The latest version supports Chinese, English, Japanese, Spanish and Arabic. You can control emotion, speaking speed and how particular words are pronounced.

It comes with a simple web page for trying it out, and there are online demos. Copying a voice is powerful, so it's best used only with voices you have permission to use.

Who it's for

Creators, researchers and developers who need natural-sounding, controllable synthetic speech.

Can I use it without coding?

With some setup. Installed with a few terminal commands, then used through a local web page; online demos are also linked. New to this? Start here.

Can I run it?

Cost
FreeNothing to pay or sign up for.
Runs on
Windows, Linux
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. Download the project and install its parts with the uv tool.
  2. Download the voice models from Hugging Face or ModelScope.
  3. Start the web demo and open it in your browser.

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. 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