AI Text to Speech

Turn English text into natural speech with 14 American and British voices, adjustable speed and WAV download.

AI Private · runs on your device Free
Runs locally in your browser AI processing happens on your device. Your files are not uploaded for inference.
AI model
…
Download size
… ·
Device
Checking browser support…
1

Kokoro speaks English (American and British voices). Text in other languages is read with English pronunciation. Up to 5,000 characters.

Natural-sounding English voices

The Text to Speech tool uses Kokoro-82M, a small open-weight speech model, to turn English text into natural-sounding speech. Use it for presentation voice-overs, video narration, language learning, rehearsal and accessibility.

Features

  • 14 American and British English voices
  • Adjustable speaking speed
  • Playback and WAV download

How it works

The text is split into sentences, synthesised on your CPU and joined into one WAV file. The model (about 90 MB) and voice files are downloaded once and cached.

Privacy and languages

Your text is never uploaded. The browser model currently supports English voices only.

How to use AI Text to Speech

  1. Type or paste your English text.
  2. Choose a voice and speaking speed.
  3. Start the synthesis and listen.
  4. Download the audio as WAV.

Why use this tool?

Produce natural English narration on your device without sending text to paid voice services.

FAQ

Does it speak other languages?

Not yet. The browser model supports English voices only; text in other languages is read with English pronunciation.

Can I use the audio in my videos?

Kokoro is an open-weight model (Apache 2.0); check the licence terms for your own project.

How long does a long text take?

Synthesis runs on the CPU, so long texts can take a few minutes depending on your device.

Related tools