Convert YouTube to Text
Sometimes you do not want a transcript with all the trimmings. You want the words: to quote a sentence in a paper, to paste a talk into a document, to read on the train, or to feed to a screen reader. This page is for that. Paste the link, turn timestamps off, copy, and you have plain text.
The rest of this page covers what to expect from the text itself: how much of it there is, how clean it will be, and how to handle the cases where captions are rough.
How much text is in a video
People speak at roughly 130 to 160 words a minute in a talk or lecture, faster in a podcast, slower in a tutorial with pauses. A ten-minute explainer is about 1,300 to 1,600 words, the length of a long article. A one-hour interview is 8,000 to 10,000 words, a novella chapter. The word count and reading time on the result page tell you exactly what you are about to copy, and the reading time is usually a quarter to a third of the video's length.
Getting clean prose, not a list of lines
Caption tracks are chopped into short cues so they fit on a screen. With the Timestamps toggle off, the tool joins those cues into flowing text before copying, so what lands in your document is paragraphs, not a stack of fragments. Uploaded captions keep their punctuation and read like an edited transcript. Auto-generated captions have almost no punctuation, so the joined text runs on; it is still searchable and quotable, but if you are pasting it into something people will read, expect to add periods.
Quoting accurately
If the point of the text is a quote, keep timestamps on for the copy, find the line, and click its time to hear it in context before you use it. Auto-generated captions sometimes get names, numbers, and homophones wrong, and a quote is the one place that matters. Copy selected text grabs only the passage you highlighted, so the quote comes over exactly as the captions have it, with its timestamp as the citation.
Where plain text goes
Into a Google Doc or Word file for a paper. Into an email to someone who would rather read than watch. Into a translation tool, which handles text far better than audio. Into a screen reader, where text is the accessible form of the video. Into ChatGPT or Claude for a summary, which the Prompts menu on the result page sets up for you. If you want the text as a file instead of on the clipboard, the TXT export is the same content with a short header naming the video, channel, source link, and date.
Other languages
The transcript is in whatever language the video's caption track is in, and that language is shown beside the word count on the result page. The tool does not translate on its own; if you need the text in another language, copy it and paste it into the translation tool you use, which is where you also get to check the result.
Frequently asked questions
How do I convert a YouTube video to text?+
Paste the link in the box above, wait a second, switch Timestamps off, and click Copy transcript. The text is on your clipboard as flowing prose.
Is there a length limit?+
No. Long videos work; the full transcript is extracted regardless of length.
Will the formatting break when I paste?+
No. The copy is plain text, so it takes on the formatting of wherever you paste it.
Can I get text from a playlist?+
One video per paste. For a playlist, work through the videos in order and export each one; the playlist guide on this site shows how to keep them together.
Do you keep my text?+
No. Transcripts are generated on demand and are not tied to you. Popular videos are cached briefly so the same captions are not fetched repeatedly.
Related tools
Try it on your video
Paste a YouTube link and get a clean transcript in seconds.