Turn interviews, voice notes, lectures and podcasts into text. Speech recognition runs privately on your device using OpenAI’s Whisper model, and you can export the transcript as TXT, Word, or SRT and VTT subtitles. Your file is processed in your browser and never uploaded to our server.
- Free to use
- No sign-up needed
- Works on phone and desktop
How it works
- Choose an audio file.
- Pick the spoken language (or detect it) and the accuracy.
- Press Transcribe. The speech model downloads once.
- Copy the text or export it as TXT, DOCX, SRT or VTT.
Good to know
- Clear recordings with one speaker at a time give the best results.
- Speakers aren’t labelled, so you’ll need to add names yourself.
- You can also translate speech into English as it’s transcribed.
Frequently asked questions
How long does it take?
It depends on your device. A 10-minute recording usually takes 2–6 minutes on a modern laptop. Fast mode is quicker but a little less accurate.
Which languages are supported?
Whisper understands about 100 languages, including English, French, Spanish, Arabic, Hindi, Chinese, Yoruba and Hausa. Accuracy is highest for widely spoken languages.
Is my audio uploaded?
No. The model runs in your browser and your recording never leaves your device.
More free tools
See all tools →🎬Convert VideoConvert videos to MP4, WebM, MOV or GIF in your browser.Open tool →🎵Convert to AudioConvert audio or video files to MP3 or WAV, and trim them.Open tool →🎞️Transcribe VideoGet a transcript and subtitles (SRT, VTT) from any video.Open tool →🎬YouTube Transcript GeneratorGet the full transcript of any captioned YouTube video, with or without timestamps.Open tool →
Use every tool without daily limitsAll Tools Pro gives you unlimited use of every tool for £9 a month + VAT. Cancel any time.
See All Tools Pro