Audio to Text
Typing up a recording takes four times as long as listening to it. This transcribes speech into text you can search, quote and edit. The speech engine runs on our own machine, so your conversation isn't handed to an outside AI service.
🔒 Uploaded for processing only — deleted automatically within 1 hour
Choose a file
Up to 100 MB · MP3, M4A, WAV, OGG · we start right away
Choose fileChoose a file — the result appears right here. Nothing else to press.
How it works
- Choose your audio or video file.
- Pick the language, or let it detect automatically.
- Download the transcript as a text file.
Good to know
FormatsMP3, M4A, WAV, OGG in — TXT out
- Writing up interview notes
- Getting a searchable record of a meeting
- Turning a voice memo into an email
Questions
Is my recording sent to OpenAI or Google?
No. The speech engine runs on our own server. Your file is processed there and deleted within the hour.
How accurate is it?
Very good on clear speech, noticeably worse with heavy background noise, crosstalk or strong accents. Treat it as a first draft you tidy up, not a court transcript.
How long can the recording be?
Up to 20 minutes per file. Transcribing takes roughly a quarter of the recording's length, so a 20-minute file takes a few minutes.
🔒 Uploaded for processing only. Your file and the result are deleted automatically within an hour — sooner once you've downloaded it.