How to convert text to speech in the browser

You can synthesize speech without uploading text to a server. Free Text to Speech loads Piper voice models from Hugging Face into your browser, runs inference with WebAssembly, and plays or downloads the result locally.

  1. Open Free Text to Speech on a modern Chromium, Firefox, or Safari browser.
  2. Paste or type the text you want to hear in the editor.
  3. Choose a voice (English or French). The first time you generate speech for a voice, the model downloads with a progress bar and is cached in IndexedDB.
  4. Adjust reading speed if needed, then click Generate and Play.
  5. Use Download WAV to save the audio file.

FAQ

Is my text sent to a server?
No. Synthesis runs on your device after models are downloaded.
Why is the first generation slow?
Voice ONNX files are large (tens of MB). Later visits reuse the cache.
Which languages are supported?
English (US) and French voices are available on this site.