Plain-AyI
← All projects
Speech & audio

ESPnet

A research toolkit for speech AI: transcription, text-to-speech, speech translation and more.

Research-backedNew this week

What it does

ESPnet is a large toolkit for building and running speech AI. It covers speech recognition (turning speech into text), text-to-speech, speech translation, cleaning up noisy audio, working out who spoke when, and singing voice synthesis.

It includes ready-made "recipes" for training models on many public datasets, plus pretrained models you can use straight away. Newer versions add a simple command line for transcribing audio.

It is aimed mainly at researchers and engineers, and training models usually needs a GPU.

Who it's for

Speech researchers, students and engineers building speech AI.

Can I use it without coding?

Developers only. It is installed with Python and used from code or the command line. New to this? Start here.

Can I run it?

Cost
FreeNothing to pay or sign up for.
Runs on
Linux, Mac, Windows
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

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.
  • License (Apache-2.0): Permissive: free to use, change and share, including for business.
  • More on staying safe

Where people are talking about it