YouTube Transcript

Where the transcript comes from

What appears after you analyse a link is not the result of listening to the audio and guessing at words. It is the caption track YouTube already serves with the video: either uploaded by the channel or generated by YouTube's speech recognition. Transcript quality therefore depends on the video, not on the tool.

That has a practical consequence. A video with captions returns text almost immediately, while a video with no track at all cannot produce text by any means. The language list you see after analysing is the complete set of what that video actually carries.

Human captions versus automatic captions

Captions uploaded by a channel usually carry punctuation, capital letters, correctly spelled names, and line breaks that follow the sentence. Automatic captions rarely have full stops or commas, often spell names phonetically, and break lines wherever the speaker paused for breath.

One detail confuses people: both kinds share the same language code. A single video can show two "English" entries in the picker — one human, one automatic. They are two different tracks, not a display glitch, and they return different text. When a human track exists, choose it first.

Why some videos return nothing

Three cases account for almost all of them. The channel turned captions off for that video, so there is no track to read. The video was published minutes ago and YouTube has not finished generating automatic captions yet — come back a little later. Or the video does carry tracks, just not in the language you picked; in that case the language list stays on screen and you simply choose another one instead of pasting the link again.

What the timestamps are for

Every segment carries a timestamp, and that is what makes reading a transcript faster than watching. You scan for the part you need, read its timestamp, and jump straight there instead of scrubbing from the beginning. On an hour-long talk this saves most of the time.

If you need a subtitle file to load into a player or an editing suite rather than text to read on the page, use the YouTube subtitle downloader — same source data, exported as TXT, SRT or VTT.

Frequently asked questions

Do I need to sign in to YouTube?

No. The tool only reads data YouTube already serves publicly alongside the video.

Do private or age-restricted videos work?

No. Those videos are not served publicly, so no caption track can be read.

Is the transcript cut short on long videos?

No. Every segment is returned; longer videos simply take a little longer to analyse.

Why is the first analysis slower than the rest?

The first request fetches data from YouTube and usually takes a few seconds. Switching language or searching the text afterwards happens in your browser and is close to instant.