How to transcribe a YouTube video
Transcrib Si transcribes files, not links: download the audio of the video, upload the file, and you get the transcript with timestamps and speaker labels a minute later. The limit is 64 MB and 120 minutes per file, the free plan is $0 with 3 files a day, and Pro is $10 a month when you pay yearly.
Step 1 — get the audio out of the video
There is no YouTube integration in the product: no link field, no downloader, no API for it. You take the audio yourself and you upload the file. With yt-dlp that is one command:
yt-dlp -x --audio-format mp3 -o "video.%(ext)s" "https://www.youtube.com/watch?v=VIDEO_ID"Any lawful downloader does the same job. Download only what you have the right to download — a lecture you may listen to is not automatically a video you may republish.
Step 2 — upload it and read it
Drop the audio file on the widget below. You get the transcript with a start and end time for every segment, speaker labels if you switch them on, and six export formats — SRT and VTT if the point is subtitles.
Step 3 — check the size before you upload
Video audio is the place where the 64 MB limit bites first. The sizes below are calculated at the given bitrate, not measured per video.
| Video length | Audio at 128 kbps | Audio at 64 kbps |
|---|---|---|
| 10 minutes | 9.6 MB | 4.8 MB |
| 30 minutes | 28.8 MB | 14.4 MB |
| 60 minutes | 57.6 MB | 28.8 MB |
| 120 minutes | 115.2 MB — over the limit | 57.6 MB |
The captions of the video, or a fresh transcript
The captions on the video were produced by somebody else's pipeline, often automatically and often without timecodes you can reuse. What you get here is a fresh transcription of the audio, with segment timecodes, with speakers separated if you ask for it, and with the text in a file you own and can edit.
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 paste a YouTube link?
No. The service takes files, not links: there is no link field in the app and no YouTube API in the product. Download the audio and upload the file.
Does it work with the captions of the video?
It does not read them. What you get is a fresh transcription of the audio with segment timecodes and, on request, the speakers separated. That transcript is yours to edit and export.
How long a video can I transcribe?
120 minutes and 64 MB per file. A two-hour video fits at 64 kbps (57.6 MB, calculated); at 128 kbps it is 115.2 MB and has to be split or re-encoded at a lower bitrate.
Can I transcribe a whole playlist?
Not from the service: one file per job. A playlist is a script — take the list, download the audio, and loop over the files through the API.
What about a live stream?
Record it first and upload the recording. There is no live transcription in the product.
Related pages.
The general guide to the service.
How to transcribe a long recording
When the video is longer than two hours or 64 MB.
The same workflow for a call recording.
Plans, limits and fair use.
Mark Burmistrov, 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.