How to get the transcript of a YouTube video
Search for this and you get a wall of sites offering to fetch a transcript for you. Most of the time you do not need any of them, because YouTube shows you the transcript itself, for free, on the page you are already looking at. What follows is the official route for someone else’s video, the official route for your own, and the honest account of when neither is good enough.
Someone else’s video: the transcript is in the description
On the desktop site, scroll down to the description under the video and expand it. At the bottom of that block there is a section headed Transcript, with a line reading “Follow along using the transcript” and a Show transcript button. Click it and the transcript opens in a panel to the right of the video, scrolling along with playback.
If you last looked for this a while ago, note that it moved. It used to live under the three-dot menu beside the video, and that is where most older instructions still send you. It is in the description now.
Two useful details once the panel is open. Clicking any line jumps the video to that moment, which makes the panel a fast way to find the one bit you actually wanted. And the three-dot menu in the panel’s own header has a Toggle timestamps option, so you can strip the times out before copying the text.
Mobile web does not have this panel at all. If you are on a phone, request the desktop site or come back to it on a computer.
When there is no transcript to show
The transcript panel is built from the video’s captions, so a video with no captions has no transcript. YouTube’s own list of reasons automatic captions might not exist is worth knowing, because it explains most empty cases:
- The captions are still processing - complex audio takes longer.
- Automatic captions do not support the video’s language.
- The video is too long.
- The sound quality is poor, or the speech is not recognised.
- There is a long stretch of silence at the start.
- Several people speak at once, or several languages are used at once.
Automatic captions are generated for 67 languages on both long-form videos and Shorts. For live streams they are English only.
Your own video: download the caption file from Studio
For a video on your own channel, YouTube will give you the captions as a file:
- Sign in to YouTube Studio.
- Pick Subtitles in the left menu.
- Click the video.
- Click Edit next to the language you want.
- Open Options and choose Download subtitles.
One wrinkle: for an untouched automatic caption track the button next to the language reads Duplicate and edit rather than Edit, because YouTube generates a new, editable track from the automatic one rather than letting you change it in place.
Whatever file you end up with, our free subtitle converter will turn it into plain text, or into a different caption format, without uploading it anywhere.
What automatic captions are actually like
YouTube is unusually direct about this. Its own documentation says automatic captions are produced by machine learning “so the quality of the captions may vary”, that creators should add proper captions first, and that automatic captions “might misrepresent the spoken content due to mispronunciations, accents, dialects, or background noise”. It closes by telling creators to review and fix them.
There is also a setting most people never see. Don’t show potentially inappropriate words is on by default, and it replaces those words in automatic captions with [ __ ]. If you are quoting from an automatic transcript and hit that, the word was not missing from the audio - it was filtered.
Then there is the shape of the text. Automatic captions are written to be read in sync with the video, in short fragments timed to speech. That is a different job from being read as a document, which is why an auto transcript pasted into a doc feels like a wall of clauses even when every word is right.
When to transcribe the source file instead
If you have the original recording - your own interview, your own podcast, your own lecture - going back to the file gives you things the caption track does not have:
- Speaker labels. Automatic captions do not tell you who is talking. For anything with two or more voices, that is usually the whole point.
- A transcript written to be read, with paragraphs, rather than caption fragments.
- Translation into another language from the same pass.
- No profanity filter deciding what you get to quote.
That is what Speecho does: upload the audio or video, get a transcript with timestamps and speaker labels, plus subtitles and a summary if you want them. The first 15 minutes are free and there is no subscription.
If what you have is a video file rather than audio, the extract audio tool will pull the soundtrack out in your browser first. And if you are preparing your own upload, the chapter formatter turns rough timestamps into a list that meets YouTube’s chapter rules.
One thing worth saying plainly
Downloading someone else’s video to get at its audio is a different act from reading the transcript YouTube publishes, and YouTube’s Terms of Service treat it that way. The transcript panel and the caption file for your own channel are both features YouTube gives you deliberately. Use those, and use Speecho for material you recorded yourself.
