MPEG-1 Audio Layer III, a lossy audio coding format
MP3 to text
Transcribe an MP3 to text
Transcribe an MP3 file to text: upload the recording, get a line per segment with timecodes, export TXT, SRT, VTT or CSV. Free $0, 3 files a day.
3 files a day up to 30 minutes, free99+ languagesSpeaker separation and timestamps
In short
Upload the MP3 as it is and get the words as text: one line per segment, the time it starts and ends, and the speakers named on request. No conversion first — a file may be up to 30 minutes, and the free plan is $0 with 3 files a day.
How to turn an MP3 into text
- 1
Drop in the MP3
The widget takes the file as the recorder wrote it. Nothing has to be converted first: the audio inside the MP3 is what the transcript is made from.
- 2
Pick the language, split the voices
Rthe recording language is detected automatically. If two people talk in the recording, switch the split by speaker on, so the lines do not end up in one column.
- 3
Read it, correct it, take it away
Every line carries the time it starts and ends. Fix a name in the app, save the transcript, and export plain text, Markdown, SRT, VTT or CSV on any plan, and DOCX on the Pro plan.
What you get
- TXT
Plain text for an email, a note or a chat
- SRT
Subtitles for video editors and platforms
- VTT
Subtitles for a browser and a website
- CSV
A table: time, speaker, text
- DOCX
A Word document with speakers and times
- Markdown
Text with headings for notes
What comes back from an MP3
commonly 75 to 95 percent smaller than CD-quality digital audio
14 in MPEG-1 Layer III, from 32 to 320 kbit/s; 128 kbit/s is the common one
ID3 tags are the usual way to carry the title, the artist and the album
audio and video are both accepted: mp3, wav, m4a, mp4, webm, ogg, flac, aac, wma, mov
The case
Why the file does not have to be converted
An MP3 is the file most people already have. Phone recorders, podcast downloads and dictation apps all write one, so the format is rarely the reason a recording cannot be turned into text. What MP3 does is compress: MPEG-1 Audio Layer III drops the parts of the sound a listener would hardly miss, and the file shrinks by 75 to 95 percent against CD quality.
Tips
Send the original file; there is nothing to convert.
A recording made on a phone is usually an M4A, not an MP3.
Read the first minute before you export the whole thing.
Switch on the split by speaker for a conversation.
Questions this page answers
Does the compression spoil the transcript?
We have not measured the accuracy of a transcript on MP3 files, so we publish no percentage for this format. What MP3 does is compress: MPEG-1 Layer III drops the sound a listener would hardly miss, and the file shrinks by 75 to 95 percent against CD quality.
Which bit rate should the file have?
Any of them. MPEG-1 Layer III allows 14 bit rates from 32 to 320 kbit/s, and 128 kbit/s is the common one on the web. We have no measured comparison between them, so we do not say which one is better.
Do the tags inside the file matter?
Not for the transcript. An MP3 may carry ID3 tags with a title, an artist and an album; the transcript is made from the sound, and the tags are not part of it.
What about a recording 30 minutes on the free plan?
Then it has to be cut. The limit of up to 30 minutes is on one file, on every plan, so a long recording goes up as separate parts.
Sources
- EWikipedia — MP3 (the .mp3 file format)en.wikipedia.org · checked 22 October 2026
- TTranscribsi — the pricing pagetranscrib.si · checked 22 October 2026
Reading: 5 minutesUpdated Transcribsi editorial team
Drop a recording — the text arrives in a minute
Speakers by name, a time on every line, export to TXT, SRT, VTT, CSV and Markdown.
3 files a day up to 30 minutes, free99+ languagesSpeaker separation and timestamps