How to transcribe audio to text in the browser
- Visit Free Speech to Text and click Load model. The multilingual Whisper model downloads once and caches locally.
- Choose the spoken language (English, French, or Swahili).
- Upload an audio/video file or use Start microphone to record.
- Click Transcribe. Word-level timestamps appear in the transcript panel.
- Copy text or download TXT/SRT exports. Nothing is sent to our servers.
Tip: Use headphones when recording to reduce echo. For long files, keep the tab active until transcription completes.