ListenKit › Guides › Free text to speech

Free text to speech online, with a natural voice

Most free text to speech sites either sound robotic or cap how much you can listen to. ListenKit runs an open neural voice on your own device instead.

How to use it

  1. Open ListenKit and paste your text, or open a PDF, EPUB or text file.
  2. Keep Natural voice on this device selected and pick one of eight English voices.
  3. Press Play. The first time, your browser downloads the voice model (about 90 MB); after that it starts quickly and works offline.
  4. Click any sentence to jump there, and drag the speed slider up to 3×.

Why it is free

The voice is Kokoro-82M, an Apache-2.0 licensed model small enough to run in a browser. Because the speech is generated on your computer or phone, we do not pay per character, so there is no daily limit to enforce. The site is supported by optional tips.

What to expect

On a laptop with WebGPU the voice is usually faster than real time. Older phones may pause briefly between sentences; the System voice option is instant on any device. There are eight English voices, and no OCR for scanned pages.

Open ListenKit

FAQ

Is there a character limit?

Each reader holds up to 500,000 characters (roughly a long novel). There is no daily limit.

Does it work offline?

Yes, once the voice model has downloaded and the page is cached, the natural voice keeps working without a connection.

Which languages are supported?

The natural voices are English (US and UK). The System voice option can use any language your device has voices for.

More guides