How do you get the transcript of a YouTube video?

The quickest way to get a YouTube transcript is to read the one YouTube already shows, when the video has one. A transcript on the video page needs no download, and you can copy its text. If the video has no transcript, you have to make one from the audio, and that is where the rules matter. YouTube's terms let you view or listen to its content for your personal, non-commercial use ¹, and they say you may not download any part of the service or its content unless the service specifically permits it ².

This page covers the route in order: the transcript on the page, the steps when there is none, and the limits that apply to each.

Start with the transcript the video page already shows

Not every video has one. When it does, you open it from the video page and read it while the video plays. Because the platform shows the text, reading or copying it is the lightest route, and it does not involve saving any media.

The steps on YouTube's own help page change over time, and we could not open that page while preparing this answer. Check the current menu on the video page, and read YouTube's help page for transcripts and captions if the menu differs from what you see. For the reasons a transcript may be missing, see why the YouTube transcript is not showing.

Copying the text is different from downloading the video

Two actions are easy to confuse. In our reading of the terms, copying text that YouTube displays to you stays within the permission to view. Saving the video file, or running a downloader to pull out its audio, is the step the terms restrict, unless the service specifically permits it ².

Read the copied text against the video before you rely on it. The platform's own captions are machine-made or human-made, and they carry their own errors. A 2017 study of YouTube's automatic captions found lower accuracy for women and for speakers from Scotland ³. Names, numbers and quotations are the places to check first.

If there is no transcript, transcribe the audio yourself

Without a transcript, the job becomes a normal transcription. You need the audio, and the terms above govern how you get it. If you own the video, or the owner gives you permission, you can transcribe your own file with any speech recognition tool.

For a free model, Whisper, the open model from OpenAI, supports 98 languages, and its accuracy varies by language ⁴. The Python package needs the ffmpeg command-line tool installed on your system ⁵. The guide to video to text explains the steps, from extracting the audio to reviewing the draft.

Review the whole draft before you use it. In a 2024 study, roughly 1 percent of Whisper transcriptions contained entire hallucinated phrases or sentences that the audio did not contain ⁶. Check quiet passages and music sections first, because that is where invented text tends to appear.

It depends on whose video it is and what you will do with the text

• Your own video. You own the recording, so the copyright question does not arise in the same way, and transcribing your own file is the simplest case.
• Someone else's video, for personal study. The terms still bar downloading. Reading the transcript the page shows you is the safer route.
• Someone else's video, for work or publication. Copyright and the terms both matter. Ask the owner for the transcript or for permission. The guide to video to text explains the copyright questions under US law, and it is general information, not legal advice.
• A video with no speech. Music or sound effects produce no useful transcript, and a model can still invent text over them.

What to do

1. Open the video and look for a transcript on the video page. If one is there, read it, and check the important passages against the audio.
2. If there is no transcript, decide whether you own the video or have permission to use it.
3. If you may not download it, ask for a transcript from the owner, or use a transcript the platform already provides.
4. If you may transcribe it, run the audio through a model such as Whisper, set the language, then check names, numbers and any silent passages before you use the text.

For a recording you already have, the free audio to text tool works in the browser. For transcripts that need timing, see captions and subtitles. For the wider subject, start with transcription. If your footage is in an editor, see how to transcribe audio in Premiere Pro or whether DaVinci Resolve can transcribe audio for free. For the term used here, see the entry for automatic speech recognition.