Convert MP3 to text
Drop an MP3 into GhostMinutes and get back a text transcript that shows who said what and when. Files can be up to 1 GB, which is roughly 18 hours of audio at 128 kbps, and you can try it without an account.
- ✓MP3 files up to 1 GB
- ✓Know who said what, with timestamps
- ✓Download TXT, DOCX, PDF, SRT or VTT
Drag & drop your audio or video files
or Browse Files
Any audio or video format (MP3, MP4, MOV, M4A, WAV, FLAC, MKV…) · Up to 1 GB · Multiple files supported
No card and no signup. Free uploads cover the first 30 minutes of a recording.
Your audio is auto-deleted after transcription. Transcripts are encrypted at rest and you can delete them any time.
How to convert an MP3 to text
- 1
Upload the MP3
Drag it into the box above. To transcribe several MP3s in one go, use the upload page in the app.
- 2
Let it transcribe
The AI detects the language and separates the speakers. Your file appears in Recent Files as soon as it is done.
- 3
Review the text
Rename speakers and check names or technical terms, which are the words speech recognition most often gets wrong.
- 4
Download the format you need
TXT for plain text, DOCX or PDF for documents, SRT or VTT for captions.
What affects MP3 transcription quality
Bitrate
Speech holds up well at ordinary MP3 bitrates, such as 128 kbps. Very low bitrates lose the consonants that separate similar words, so use the highest-quality copy you have.
Background music
Music under speech makes transcription harder. If you have a version of the track without music, upload that one.
Several people talking at once
Speaker labels work best when people take turns. Overlapping speech is the most likely part of a transcript to need a manual fix.
Which export to pick
TXT or DOCX
Best for quotes, notes and editing the text further.
Best for sharing a finished transcript that should not be edited.
SRT or VTT
Time-coded caption files for video players and editors, useful when the MP3 is the soundtrack of a video.
What you get with GhostMinutes
No bot in your call
You upload the file after the call, so nothing shows up in the participant list.
Old recordings work too
Upload a recording from today or from last year. GhostMinutes takes MP3, MP4, MOV, M4A, WAV, FLAC, MKV and other audio or video formats up to 1 GB.
Speaker labels and timestamps
Each line shows who spoke and when. Rename the speakers, then ask the AI questions about what was said.
95+ languages
Transcribe and summarize in more than 95 languages. GhostMinutes detects the language of each file automatically.
Audio deleted after transcription
The web app deletes your audio file right after processing. Transcripts are encrypted at rest and kept for up to 90 days unless you delete them sooner. We never sell them, share them or train on them.
Free plan, then from $1 a month
Free includes 150 minutes a month. Lite is $1 a month for 600 minutes, and Pro is $5 a month for 3,600 minutes.
MP3 to text questions
Try it on your own recording
Upload a file and get the transcript, speaker labels and an AI summary. No card and no signup.