Which languages does it understand?

English. Recordings in other languages come out as nonsense, so they aren't a good fit yet.

How long can a recording be?

Up to an hour. The recording is split at natural pauses and transcribed piece by piece, so you see the progress as it goes; a long file takes a few minutes.

Does it tell speakers apart?

No — the transcript is one continuous text. With timestamps on, every pause starts a new paragraph with its time, which makes it easy to find a moment in the recording and mark who spoke.

Audio to Text

Transcribe a recording to text — interviews, lectures, meetings, voice memos. Save it as a text file or a PDF, with timestamps if you like.

Zero-transmission processing
No watermark
Limited daily trial
Unlimited on Pro

Drop your MP3, M4A, AAC, WAV, OGG, Opus, WebM, FLAC, MP4 here

or choose it from your files

Your files are never transmitted. Nothing is uploaded.

PDF converter54 ready

Good to know

Accepts: MP3 · M4A · AAC · WAV · OGG · Opus · WebM · FLAC · MP4
Files at once: One at a time
Transmission: None — nothing is uploaded.

Chain it

Run this tool and the next one back to back — nothing to download in between.

Build a workflow

How to do it

Add your file
Add the recording — MP3, M4A, WAV, OGG, Opus, FLAC, WebM, or an MP4 video. Phone voice memos work as they are.
Set it up the way you want
Choose a text file or a PDF, and whether to mark each paragraph with the time it starts.
Download the result
Download the transcript. Read it through once — names, numbers and rare terms are where mistakes happen.

Questions about Audio to Text

Which languages does it understand?: English. Recordings in other languages come out as nonsense, so they aren't a good fit yet.
How long can a recording be?: Up to an hour. The recording is split at natural pauses and transcribed piece by piece, so you see the progress as it goes; a long file takes a few minutes.
Does it tell speakers apart?: No — the transcript is one continuous text. With timestamps on, every pause starts a new paragraph with its time, which makes it easy to find a moment in the recording and mark who spoke.