Free up to 30 minutes · no sign-up
Upload an MP3 and get its words back as text, split into sentences, labeled by speaker and stamped with the time each sentence starts. Files up to 30 minutes are transcribed in full for free.
Transcribe files
Drop audio or video files here
MP3, M4A, WAV, FLAC, OGG, MP4, MOV and more · up to 3 files at once
Browse filesNo account needed: 3 files a day free, full transcripts up to 30 min each. for 30 more free minutes.
Drop one MP3 or several into the box above, or tap Browse files on your phone. Up to 3 files at once without an account, 50 with one.
Leave it on Detect automatically, or choose the language if the recording starts with music or silence. Speaker labels are on by default.
The transcript appears below the box. Copy it in one click, or download it as TXT, Word (DOCX), SRT or VTT.
| Without an account | Free account | |
|---|---|---|
| Free files per day | 3 | 3 |
| Files up to 30 minutes | Full transcript | Full transcript |
| Files over 30 minutes | First 10 minutes | First 10 minutes, or use your minutes |
| Extra minutes | None | 30 on sign-up |
| File size | 500 MB | 2 GB |
| Files at once | 3 | 50 |
| TXT, Word, SRT and VTT downloads | After a free sign-up | Yes |
| Credit card | No | No |
Every file: up to 10 hours, 90+ languages, speaker labels and timestamps. After that, one-time packs: 3 hours for $6.99, 15 hours for $19.99 or 50 hours for $49.99, and minutes never expire. The original audio is deleted after transcription (within 48 hours at most), and we never use your files to train AI models.
An MP3 is just compressed audio, so the converter listens to it the same way it would a WAV or a voice memo. What comes back is not a single block of words: the text is split into sentences, punctuated, and grouped by who is speaking, with Speaker A, Speaker B and so on. Filler sounds like "um" and "uh" are left out, so the text reads cleanly.
Every sentence carries the time it starts in the MP3. That makes it easy to jump back to the recording and check a quote, or to turn the transcript into subtitles: the SRT and VTT downloads use the same timings.
Bitrate matters much less than people expect. Speech is understood well even at 64 kbps or in mono, which is what most recorder apps, podcast downloads and call recorders produce. You do not need to convert an MP3 to WAV first; that only makes the file bigger.
What does matter is how the speech was recorded. Distance from the microphone, echo, music under the voices and people talking over each other all lower accuracy. If a recording is hard for you to follow, expect the transcript to need a closer check too.
| MP3 source | Typical quality | Tip |
|---|---|---|
| Podcast episode | Very good | Turn speaker labels on, set the number of hosts if you know it |
| Phone call recording | Good | Pick the language; calls are often short and start with a ring tone |
| Lecture recorded from the back of a room | Fair | Turn speaker labels off if only the lecturer matters |
| Meeting on a laptop microphone | Fair to good | Set the number of speakers for cleaner labels |
Choose the download that fits what you will do next. Word (DOCX) keeps the speaker names and timestamps and is the easiest to edit, share and comment on. TXT is plain text for notes, search or pasting into another tool. SRT and VTT are subtitle files you can upload to YouTube, Vimeo or a video editor alongside the original audio or video.
Most MP3s people convert are short: a voice note, a call, a single interview. Those are free: 3 files a day, full transcripts up to 30 minutes each, with no account. Longer MP3s show their first 10 minutes free, and you can unlock the rest with minutes. Minutes come in one-time packs, from $6.99 for 3 hours, and they never expire.
Yes for MP3s up to 30 minutes: you get the full transcript free, 3 files a day, without an account. Longer files show their first 10 minutes free. Signing up adds 30 free minutes, and after that minutes come in one-time packs from $6.99 for 3 hours.
Up to 2 GB and 10 hours per file with a free account, or 500 MB without one. A one-hour MP3 at 128 kbps is about 60 MB, so most recordings fit easily.
Yes. Add up to 3 MP3s at once without an account, or up to 50 when signed in. They are transcribed one after another and each appears below the upload box when it is ready.
Yes. Download the transcript as DOCX and open it in Word, Google Docs or Pages. Speaker names and timestamps are kept. Downloads need a free account, so a transcript made without one moves to your account when you sign up.
Yes. It runs in the browser, so you can pick an MP3 from Files on iPhone or from your Downloads on Android without installing anything.
Usually a few minutes for an hour of audio, and well under a minute for a short voice note.
Yes. Without an account you can convert 3 MP3s a day and read or copy the full text of any file up to 30 minutes on the page. Downloading TXT, Word or subtitles needs a free account.
More than 90, detected automatically. Pick the language yourself if the MP3 is very short or starts with music or silence.
No. The MP3 is deleted after transcription, within 48 hours at most, and we never use your files to train AI models. The text stays in your account until you ask us to delete it.
Files up to 30 minutes get a full transcript free, no account needed.
Upload a fileUpdated October 2026