Audio to text
Transcribe speech with Whisper, on your device.
Runs entirely in your browser. Verify it in your network tab.
What this does
Runs OpenAI Whisper (tiny.en) locally through WebAssembly. The model is downloaded once, around 39 MB, then cached by your browser for every later run.
Limits
- English only. This is the `tiny.en` model.
- First run downloads roughly 39 MB of model weights. After that it is offline.
- Accuracy is well below the full-size cloud models, especially with accents, crosstalk or background noise.
- Long recordings are slow: expect a sizeable fraction of the audio duration.
Is this private?
Yes, and you can check rather than take our word for it. Open your browser's network tab and run the tool: your file never appears in it, because there is no server to send it to. The whole site is static files, plus the WebAssembly engine that does the work, which is served from this same domain.