Browse projects/Audio and music/Turn text into speech/kokoro
hexgrad/kokoro A tool that works on your files
Read English text aloud in a chosen voice and at a chosen speed
- Open source
- Runs on your own computer
- Windows, Mac or Linux
- Tested 5 Nov 2026
You bring
- English text in a UTF-8 text file
You get
- a spoken recording of your text
What you can change
Set these in UGT before you start.
- English voice; defaults to af_heart
- reading speed as a multiplier from 0.5 to 2; default 1
What can it reach?
Only the files you give it, with no network.
How was it tested?
Run on clean Intel and AMD and Apple silicon and ARM machines on 5 Nov 2026. Its check passed on a known sample and rejected deliberately broken results.
This checks the job this page describes, not every feature of the original project.
Project, licence and recipe files
Licence. Apache-2.0 (Kokoro and Kokoro-82M); espeak-ng GPL-3.0-or-later
Recipe revision 79dde58b18c8, from 1c7bdd971d2e.
Other ways to turn text into speech
Audio and music
Transcribe, separate, clean up and voice your audio.
