MP3 to Text
Get a written transcript of an MP3: a podcast episode, an interview, a lecture or a recorded call.
or drop them here · up to 50 MB per file, free
Bigger files? Pro converts files up to 500 MBOver the free limits: you can unlock the download from €1.99.
Working on it…
Your file is ready
Free · No sign-up · No watermark · Your files never leave your device
MP3 to Text in 3 steps
- Select one or more MP3 files, or drop them on the box.
- Choose the language or let it be detected, then click Transcribe.
- Copy the transcript, correct it, or download it as TXT or SRT subtitles.
From an MP3 recording to a transcript
A transcript makes a recording searchable and quotable: you can find the minute where something was said, copy a quote for an article, send minutes of a meeting or study a lecture without replaying it. Select the MP3 and the tool writes out what is said, with punctuation and a new paragraph at each pause. Besides the plain text you can download an SRT file, which has the same text split into short lines with start and end times, for a video editor, YouTube or a media player.
The MP3 is decoded and transcribed in your browser with Whisper, a speech model trained on hundreds of thousands of hours of audio. Bit rate matters less than you might think: a 64 kbps voice recording transcribes as well as a 320 kbps one, because what counts is a clear voice. Podcasts with music beds and jingles work, but the music itself is skipped. For a recording longer than an hour, split it first with our Audio Cutter if your device is slow; otherwise just leave the tab open until it finishes.
Frequently asked questions
How do I convert MP3 to text?
Select the MP3 file, leave the language on Detect automatically and click Transcribe. When it finishes, the transcript is shown on the page: copy it, edit it, or download it as a TXT file or as SRT subtitles with timings.
Can I get timestamps?
Yes. Download the SRT file: every line of the transcript comes with the time it starts and ends in the recording. SRT opens in any text editor and is accepted by YouTube, Premiere, DaVinci Resolve, VLC and most video players.
Does it tell speakers apart?
No. The transcript is the words in order, with a new paragraph at each pause, but it does not label who is speaking. For an interview, read through the text and add the names where the speaker changes.
How long does it take?
It depends on your device and the setting. On a recent computer with the Fast setting, a minute of MP3 takes about 20 seconds; Most accurate takes about a minute, and phones take two to three times as long. The first time, the speech model is downloaded once (about 80 MB) before transcription starts.
Are my MP3 files uploaded?
No. The MP3 is transcribed inside your browser on your own device, so the recording is never sent to us or stored anywhere. Only the speech model is downloaded from this site, once.
Related tools
Part of: Audio tools