Free text to speech with natural AI voices

How to use it

1. 1

   Paste your text. Up to 5,000 characters at a time. Shorter sentences sound most natural.

2. 2

   Pick a voice and speed. Choose from eight US and British English voices.

3. 3

   Create and download. Press Create speech. The first time, the voice model downloads (about 90 MB). Then play it or save it as a WAV file.

Can I use the audio in my videos?

The Kokoro voice model is released under the Apache 2.0 license, which allows commercial use, so you can use the audio in YouTube videos, ads and courses. As with any AI voice, don't use it to impersonate a real person.

Tips for natural-sounding speech

• Use punctuation: commas and full stops create natural pauses.
• Write numbers the way they should be said ("twenty twenty-six" vs "two zero two six").
• Spell unusual names phonetically if they come out wrong.
• Break long scripts into paragraphs and generate them one by one.

How it works

Most text-to-speech sites charge by the character and send your text to their servers. This tool runs Kokoro-82M, a small but high-quality open-source voice model, in your browser. After a one-time download, it works offline, and there's no character limit beyond 5,000 per run.

Limits

Voices are English only (US and British accents). Generation speed depends on your device; a minute of audio typically takes from a few seconds on a modern laptop to a minute or so on an older phone.

Questions