Written by the founder of BreezeConvert — a creative director working across set design, game 3D, Unreal/VFX and AI content pipelines. More about the author.
Video is a terrible format for information you actually need to use. You can't search it, quote it, skim it, or paste it into an AI assistant. So a common task is turning a YouTube video into plain text. There are three reliable ways to do it, and one popular way that quietly isn't as private as it looks.
Method 1: YouTube's built-in transcript (fastest)
Most YouTube videos already have captions, and YouTube will hand them to you. Under the video, click "…more" and then "Show transcript." A panel opens with the whole spoken text, timestamped. Select it all, copy, and you have the transcript. This is the quickest route and needs no extra tools — the only catch is that the copied text is cluttered with timestamps and, for auto-captions, repeated lines. That's easy to clean (more below).
Method 2: a caption file (.srt/.vtt)
If you've downloaded a video's captions, or you have subtitle files from an editor, they arrive as .srt or .vtt files. These are just text with cue numbers and timestamps around the words. They're the cleanest source because they're structured, and they convert to readable text perfectly.
Method 3: speech-to-text (when there are no captions)
Some videos have no captions at all. Then you need actual transcription: extract the audio (our MP4-to-MP3 tool does this locally if you have the file) and run it through a speech-to-text service. This is more work and less accurate on poor audio, so use it only when methods 1 and 2 aren't available.
Why "just paste the YouTube link" isn't as clean as it seems
You'll find lots of tools that promise "paste a URL, get the transcript." Here's the honest mechanics: a web page in your browser cannot fetch another site's captions directly — YouTube's caption endpoints block cross-origin requests for exactly this reason. So any "paste a link" tool is really sending that link to its own server, which fetches the captions on your behalf. That's fine for a public video, but it means a server is in the loop. If you value keeping things on your own machine, working from the transcript text or the caption file avoids that entirely.
Cleaning it up for reading or AI
However you get the raw text, it'll have timestamps, cue numbers, and (for auto-captions) duplicated rolling lines. Our subtitles-to-Markdown tool strips all of that and gives you clean, paragraph-grouped text — in your browser, nothing uploaded. Drop a .srt/.vtt file in, or paste the text you copied from YouTube's transcript panel, and you get readable prose. From there it goes straight into the Markdown tool to count tokens, apply a "summarise this lecture" prompt, or split a long transcript into chunks for Claude or ChatGPT.
Related guides: Extract audio from a video · Preparing documents for AI · Image / scan to text.