Transcribe MP4 to text
An MP4 is a video file, and the audio is inside it. Send the MP4 and you get the same transcript as for an audio file: the text with a start and an end time on every line, speaker labels on request, and SRT or VTT subtitles you can use on the video. But a file over 64 MB is refused, so a long video usually has to lose its picture first.
What you get from an MP4
| What | Value | Where it comes from |
|---|---|---|
| The file we take | mp4, webm and mov, together with every audio container | Service contract |
| Size of one file | 64 MB | Service contract |
| Length of one file | 120 minutes (two hours) of audio | Service contract |
| What comes back | the text, the times of every segment, the speaker labels on request | Service contract |
| Export formats | 6 — TXT, Markdown, SRT, VTT, CSV, DOCX; SRT and VTT are the subtitles | Pricing page |
| Free plan | $0, 3 files a day, files up to 30 minutes, no card | Pricing page |
| Pro | $10 per month billed yearly ($120 a year), or $20 per month billed monthly | Pricing page |
| Pro allowance | 1,500 minutes of audio per calendar month | Pricing page |
| Speed | 0.84 s to transcribe a short file, 0.39 s to detect its language | our own measurement |
The size limit comes first
A video is a picture plus an audio track, and the picture is what makes the file big. The sizes below are calculated from the bitrate of the audio track, not measured on a file.
| Recording | Size (calculated) | Fits 64 MB? |
|---|---|---|
| 10 minutes at 64 kbps | 4.8 MB | yes |
| 60 minutes at 64 kbps | 28.8 MB | yes |
| 120 minutes at 64 kbps | 57.6 MB | yes |
| 10 minutes at 128 kbps | 9.6 MB | yes |
| 60 minutes at 128 kbps | 57.6 MB | yes |
| 120 minutes at 128 kbps | 115.2 MB | no — the file is refused |
So a two-hour lecture in a 720p video does not fit, while the same two hours of audio at 64 kbps does. That is the whole reason to take the audio out first.
How to take the audio out of a video
This is a command on your own computer; ffmpeg is a free tool and the service does not run it for you. The first command keeps the audio as MP3, the second writes plain audio that the engine takes as it is.
ffmpeg -i lecture.mp4 -vn -acodec libmp3lame -b:a 64k lecture.mp3ffmpeg -i lecture.mp4 -vn -ar 16000 -ac 1 -c:a pcm_s16le lecture.wavHow to transcribe an MP4 file
-
Upload the video, or the audio taken out of it
Drop the file on the widget below, or send it with
POST /v1/transcriptsand a multipartfilefield. -
Set the language, or let the service detect it
Russian or English, or automatic detection — 0.39 s on a short file in our own measurement.
-
Ask for speaker separation
Switch on “Split by speaker” so the voices in the video are told apart.
-
Export the subtitles
Take SRT or VTT: the start and the end time of every segment are written into them, so the subtitles line up with the picture.
What we do not do
We do not cut video, we do not burn subtitles into the frames and we do not return a video file. The picture is not read: the recognition runs on the audio track only, so a slide on the screen is not in the transcript.
Put a file on the page and read the first lines.
The widget below is the same one as on the home page: file → transcript with timecodes and speakers → export. The transcript is saved to your account.
Drop an audio or video file
or — mp3, wav, m4a, mp4, up to 64 MB and two hours
Transcription runs under your account: sign in or create a free account.
No account yet? The free plan needs no card. Create a free account or sign in.
Questions this page answers.
Can I upload a video straight to the site?
Yes, if the file is under 64 MB. mp4, webm and mov are accepted, and a file over the limit is refused with a clear error instead of a silent failure.
Do you return subtitles for the video?
SRT and VTT are two of the six export formats, and the start and the end time of every segment is written into them. We do not render them into the picture and we do not return a new video file.
What about a YouTube video?
The service takes files, not links. Download the audio yourself and upload the file — there is a step-by-step guide on the page how to transcribe a YouTube video.
How long can the video be?
120 minutes of audio per file. In practice the size limit comes first: a two-hour video in a 720p container is far over 64 MB, while two hours of audio at 64 kbps is 57.6 MB.
Does the picture help the transcript?
No. The recognition runs on the audio track only, so what is written on the screen — slides, a shared spreadsheet, a name in the corner — is not in the transcript.
Can I transcribe a recording of a conference call?
Usually, yes: a conference recording is an MP4 with the audio in it. If the file is over 64 MB, take the audio out first, as above. The screen sharing is not transcribed.
Related pages.
The service takes files, not links: the three-step way.
The audio file after you have taken it out of the video.
A checklist for calls, with speakers, timecodes and exports.
Every plan, every limit, in one table.
Upload with drag and drop, the queue and the exports.
The live answer base: 99 questions about the price, the limits, the languages, the exports, the privacy and the API, each with the page of this site it comes from.
Mark Vesyoly, Founder.
Every number on this page comes from the service contract, the pricing page or our own measurements. The author of the page and the publisher of the site are the same person. Questions and corrections: support@transcrib.si.