QuietSteno

About QuietSteno

QuietSteno is a free tool for turning speech into text. It runs speech recognition inside your web browser, so recordings stay on your device.

What it is

QuietSteno converts audio and video files into editable transcripts with timestamps, and exports them as TXT, SRT or VTT. It supports 27 spoken languages and can translate speech in those languages into English text.

The name comes from the stenographer’s pad: a quiet, reliable way of getting spoken words onto paper.

How it works

When you choose a file, your browser decodes its audio and passes it to Whisper, an open-source speech recognition model released by OpenAI under the MIT licence. The model runs on your own hardware through the open-source Transformers.js library, using your graphics chip where the browser supports WebGPU and your processor otherwise.

Because no server does the transcription, there is no per-minute cost, which is why the tool can be free without minute limits.

What it does not do

  • It does not identify speakers.
  • It does not summarise.
  • It does not transcribe from a link; the file has to be on your device.
  • It is not as accurate as the largest cloud models on difficult audio. Always proofread before you publish or quote.

Privacy

Your recordings and transcripts are never uploaded. The site has no accounts and sets no cookies. Details are on the privacy page.

Credits

  • Speech recognition: Whisper by OpenAI, converted for the web by the ONNX community on Hugging Face.
  • In-browser inference: Transformers.js and ONNX Runtime Web.
  • Typefaces: Bricolage Grotesque and Atkinson Hyperlegible, both under the SIL Open Font License.
  • Sample clip: the opening of John F. Kennedy’s 1961 inaugural address, a public-domain United States government recording.

Transcribe a file

Last updated 13 November 2026.