Plain text
YouTube to Text
Turn a video into something you can actually read and edit: paragraph after paragraph of clean text, with the timestamps left out of the way.
Paste a YouTube video link
One video, or several pasted together. Duplicates are removed.
Press Ctrl + Enter to fetch
Nothing processed yet
Paste a YouTube link above and press Fetch transcripts. Each video comes back as its own card.
Any link shape works
Videos, Shorts, playlists, channels, or a bare video ID. Duplicates are removed.
Real captions, not guesses
Human-written captions are preferred, with YouTube's automatic ones as the fallback.
Copy or download
Copy as clean prose or with timestamps, or export TXT, SRT, VTT and JSON.
Playlists and channels
Process a whole playlist or channel in the background and watch progress as it goes.
Text, not a subtitle file
Captions arrive as a stream of short cues a few words long, each with its own start time. Dropped straight into a document they are close to unreadable. The Clean view joins the cues back into sentences and starts a new paragraph wherever the speaker paused, so you get prose rather than a transcript-shaped wall of fragments.
The exact same text is available as a .txt download, with no timing information in it at all.
What people use it for
- Notes and study — read a lecture or talk at your own pace, and quote it accurately.
- Summaries — paste the clean text into a summariser or an AI assistant without the timestamps confusing it.
- Translation — a tidy block of prose translates far better than a cue-by-cue transcript.
- Writing and research — search for the phrase you half-remember and jump straight to it.
- Accessibility — a text version of a video for people who cannot or would rather not watch it.
How to do it
- Leave Default view set to Clean text — that is the plain-text view.
- Paste the video link above and press Fetch transcripts.
- Press Copy on the card to put the whole thing on your clipboard.
- Or use the download menu to save it as a .txt file.
Switching to the Timestamps view is still there whenever you need to find the exact moment something was said, and switching back is instant.
Limitations
Transcripts are read from the captions YouTube already publishes for a video. This tool does not download video or audio files, and it does not run speech recognition — so a video with no captions of any kind cannot be transcribed here.
Because the text comes from automatic captions when no human captions exist, it will not have punctuation exactly where a person would put it, and proper nouns are sometimes wrong.
Frequently asked questions
- How is this different from the transcript tool?
- It is the same caption data, presented as readable prose with the timestamps removed. The transcript tool puts the timestamps front and centre; this one is for when you want text you can read and edit.
- Will the text have punctuation?
- If the video has human-written captions, yes. Automatic captions have limited punctuation and no reliable sentence boundaries, so the paragraphs follow the speaker's pauses instead.
- Can I paste the text straight into an AI tool?
- Yes, and it works better than a timestamped transcript because there is no repeated timecode noise breaking up the sentences.
- Is there a character or word limit?
- No. A long video produces a long transcript, and nothing is truncated on the way out.
Diagnostics — when a video will not load
Runs each retrieval method against one video and reports exactly what YouTube said. The fastest way to tell a blocked network apart from a video that simply has no captions.