MP3 to Text - Upload Once, Get Word-Level Timing
MP3 to text online: automatic language detection across 90+ languages, a transcript with word-level timestamps, and an SRT file. 200MB per upload.
最終更新: 2026-10-18
MP3 to text: one upload, three things back
MP3 to text means turning a recording into words you can edit, search and quote. Upload an MP3 here and you get a transcript, word-level timestamps, and an SRT file. Not three operations — three outputs of a single upload.
Most transcription tools give you only the first. The second is where the time actually goes: with per-word timing, subtitles need no alignment pass, editing needs no scrubbing to find the cut, and long audio becomes searchable by word.
Real output, errors included
Tested on a Mandarin storytelling clip from this site. Below is the unedited response:
Interviews are the most common recordings people bring to MP3 to text conversion. Long sentences, pauses and spoken word order all test sentence segmentation:
Detected language: zho (confidence 88.7%)
Transcript:
上回书说到，那少年提剑立于城门之西，身后是三千追兵，
身前是莫张深渊。他回头笑了一声：“诸位，且听下回分解。
Word timings (first 12):
上[0.42→0.72] 回[0.72→0.88] 书[0.88→1.06] 说[1.06→1.48]
到[1.48→1.84] ，[1.84→2.62] 那[2.62→2.96] 少[2.96→3.18]
年[3.18→3.40] 提[3.40→3.64] 剑[3.64→3.96] 立[3.96→4.22]
It got two things wrong. The source reads 城门之下 and 万丈深渊; the engine heard 城门之西 and 莫张深渊 — homophone substitutions.
Homophones are a hard limit for every speech engine. We leave the errors on the page because that is more useful than a hand-picked perfect sample: you now know what to expect. Most words are right; proper nouns and set phrases need one pass of your eyes.
The important part: fixing text does not move the timings. Correct the word in the studio and its start and end times are unchanged, so the subtitles still line up.
What word-level timing is actually for
| Task | Without timings | With them |
|---|---|---|
| Producing subtitles | Re-transcribe inside the editor, then align by hand | Export the SRT, drop it in, done |
| Finding a specific line | Scrub the playhead back and forth | Search the text, click, jump |
| Cutting a passage | Audition repeatedly for the in and out points | Select the words, read the exact seconds |
| Clipping for social | Cut by feel | Cut on sentence boundaries, never mid-word |
The first row saves the most time. The SRT is computed during transcription, not transcribed again afterwards — so you never end up with one set of typos in the subtitles and a different set in the transcript.
Speaker separation is on by default
If more than one person is speaking, the system separates them automatically and labels who said what, up to 32 speakers. Interviews and two-host podcasts come back already shaped as dialogue.
See meeting transcription for how that performs and where it struggles.
MP3 to text works for other audio formats too
The page says MP3 to text, but MP3 is only one of the accepted formats. WAV, M4A, AAC, OGG and FLAC upload the same way, up to 200MB total per upload.
Upload whatever your recorder, phone or podcast software exported. There is no need to convert to MP3 first — converting only adds another round of encoding and does nothing for accuracy. An hour of stereo MP3 is usually 60–120MB, so most interviews, episodes and lectures fit in one upload.
Working from a video file instead? MP4 to text takes video directly, with no audio extraction step.
How to convert MP3 to text
- Open the voice studio → transcribe
- Upload the file (MP3/WAV/M4A/AAC/OGG/FLAC, 200MB total per upload)
- Wait — an hour of audio is typically one to two minutes
- Fix any wrong words, then export the transcript or the SRT
No language to select, and speaker separation and audio-event tagging are on by default.
Four common mistakes
- Converting the format before uploading. Upload the original file; transcoding only costs quality.
- Exporting after a glance. Ordinary words are mostly right, but names, places and idioms are where homophones hide — check those specifically.
- Fixing subtitles cue by cue in the editor. Correct the transcript in bulk first, then export the SRT. The timings do not move, so you proofread once.
- Editing the audio after transcribing. Cut a passage and every later timestamp shifts. Lock the edit, then transcribe.
FAQ
Which audio formats work, and how large can the file be? Anything the browser recognises as audio — MP3, WAV, M4A, AAC, OGG, FLAC. The per-upload limit is 200MB total, and an hour of stereo MP3 is usually 60–120MB, so most interviews, podcast episodes and lecture recordings fit in a single upload.
How accurate is MP3 to text conversion? Clean single-speaker audio usually lands above 95%, but homophones remain a hard limit for every speech engine — the real output shown on this page contains two such errors. We publish the unedited result because you can fix words directly in the studio, and editing text does not disturb the timings already computed.
Do I have to select the language first? No. The language is detected automatically and returned with a confidence score, and mixed-language audio is handled without being told in advance. If you already know the language you can specify it for a small accuracy gain.
Can the transcript be exported as subtitles? Yes. Every transcription returns start and end times per word, which export directly as an SRT file. Dropped into CapCut, Premiere or Final Cut it lines up with the audio without a second alignment pass.
Start here
Cost: 20 credits per file (up to 200MB per upload), refunded automatically if it fails. The $3.90 trial pack has 150 credits, enough for 7 files, and they never expire.
Pick the recording you most need written up and run it through MP3 to text — then see how much re-listening the word timings save you.
Related: video to text · meeting transcription · subtitle generator · speech to text overview