Audio
Speech to Text Transcriber
Turn a recording into text with time stamps. It runs in your browser, so the sound never leaves your device.
Preparing tool
How to use Speech to Text Transcriber
- Choose an audio or video file, or try the sample.
- Press Start. The first time, the speech model downloads (about 41 MB) and is kept by your browser.
- Fix any words, click a time to hear it, then download TXT, SRT or VTT.
What are the limits of Speech to Text Transcriber?
English only (the small English Whisper model). Accuracy drops with noise, strong accents and several people talking at once, and there is no speaker labelling. Time stamps are about one second accurate. It runs on your device, so long files are slow on phones; keep files under 60 minutes. Nothing is uploaded.