Answers
What does TranscriptsDisabled mean, and can I still get the YouTube transcript?
Short answer
TranscriptsDisabled means the library found no caption track it could read for that video. Ask a second source to tell a truly captionless video from a blocked request: the YouTube transcript endpoint returns the text for $0.004, or not_found at no charge when the video really has no captions.
You run a batch of 200 videos and 30 of them come back with TranscriptsDisabled. Some of those videos really have no captions. Others show a CC button in the browser, and people in the youtube-transcript-api tracker report exactly that case. You need to know which is which before you decide what to do with them.
Two different causes
- The video has no caption track. The creator turned captions off, YouTube has not generated automatic captions yet (new uploads can take a while), or the audio is music with no speech. No tool can return a transcript that does not exist.
- The request was answered differently. When YouTube treats your IP as suspicious, the page your code receives can look like a video without captions. The error says disabled, but the cause is the request.
A cheap second opinion
Send the same video to YouTube video transcript. The fetch does not come from your IP, so the answer reflects the video itself:
curl -s -G "https://api.okdata.online/v1/youtube/videos/transcript/get" \
-H "Authorization: Bearer $OKDATA_API_KEY" \
-d id=VIDEO_ID -d fields=lang,text
A video without captions answers like this, and the charged: false is real:
{ "code": "not_found", "message": "No YouTube transcript with that ID was found.", "request_id": "…", "charged": false, "retry": "no" }
| Outcome | What you pay |
|---|---|
| Transcript returned | $0.004 |
not_found, no captions |
nothing |
| Any error (rate limit, timeout, upstream) | nothing |
So re-checking 30 failed videos costs at most $0.12, and less for every one that truly has no captions.
What to do with the truly captionless ones
- Recent uploads: try again in a day. Automatic captions often appear later.
- Talks and podcasts: run Whisper or another speech model on audio you are allowed to download. That is the only way to get words out of a video with no track.
- Tutorials: look at the description. Many creators paste a script or chapter list there; YouTube video detail returns it as
textfor $0.004. - Your own channel: upload captions in YouTube Studio, then fetch them again.
Keep the batch honest
Store the outcome per video: ok, no_captions or error. Retry the errors, schedule the new uploads for later, and send only the rest to speech-to-text. The response never says whether a track was typed by the creator or generated automatically, so treat automatic-caption accuracy with care when you quote.
Batch patterns and costs for whole lists are on the video transcript solution page; every error code is on errors.
Updated