MP3 to text, in seconds
MP3 is the format everything outputs — voice recorders, podcast tools, call apps. Upload it here and get clean, punctuated text back. Free to start.
Every recording device speaks MP3. Old dictaphones, podcast recorders, Zoom's separate-audio export, that voice memo someone emailed you in 2019 — if it ends in .mp3, it works here.
Upload the file and Whisper transcribes it with punctuation and paragraphs, in 90+ languages. The 25 MB limit covers about two hours of typical spoken-word MP3.
What people convert
Show notes, quotes, and searchable archives from your own recordings.
Doctors, lawyers, and field workers turning spoken notes into records.
Sales calls and support lines, reviewed as text instead of replayed.
Questions, answered
How big an MP3 can I upload?
25 MB per file — roughly two hours of spoken audio at typical bitrates.
Does it work with stereo files?
Yes. Channels are mixed automatically; stereo and mono MP3s both transcribe the same.
MP3 is lossy — does that hurt accuracy?
Barely. Whisper was trained on compressed, real-world audio; a standard 128 kbps MP3 transcribes as well as a lossless original.
What about other formats?
m4a, wav, ogg, webm, flac, mp4 audio and more — see the format pages linked below.