How to get a YouTube transcript for a video you own
Download your own upload from YouTube Studio, transcribe the MP4 in your browser, and turn the result into a YouTube transcript, show notes or captions.
This is about a video you uploaded yourself. YouTube Studio will give you an MP4 copy of your own upload: open Content, pick the video, then select Menu and Download. Run that MP4 through a transcriber that works in the browser and you have a YouTube transcript in a few minutes, with the file never leaving your computer. Downloading a video you do not own is a different matter, and YouTube’s Terms of Service do not allow it.
Get the MP4 out of YouTube Studio
Google documents the route on its help page for downloading videos you have uploaded. Sign in at studio.youtube.com, select Content in the left panel, find the video, then select Menu and Download.
What lands on your disk is an MP4, at either 720p or 360p “depending on the video size”. It is not your original master file, and for transcription that does not matter, because only the audio track gets read.
Google also lists the cases where the Download option will not work for you:
- The video has been removed from YouTube.
- It has copyright or Community Guidelines strikes.
- It uses preapproved audio tracks.
- You have already downloaded it 5 times in the last 24 hours.
That last one is the rate limit: 5 downloads per video per 24 hours, after which the option comes back. And the same help page states the boundary plainly: “You can’t download other user’s YouTube videos.”
Someone else’s video is not yours to download
The Terms of Service say the same thing in wider language. Under Permissions and Restrictions, you are not allowed to “access, reproduce, download, distribute, transmit, broadcast, display, sell, license, alter, modify or otherwise use any part of the Service or any Content except: (a) as expressly authorized by the Service; or (b) with prior written permission from YouTube and, if applicable, the respective rights holders.”
The Download button inside your own Studio account is the Service expressly authorising it. A third-party downloader aimed at a channel you do not run is not, and this guide will not tell you how to use one. If you need the words from someone else’s video, ask the creator.
What YouTube’s automatic captions already do
Before you transcribe anything, check what YouTube has generated on its own. Automatic captions are produced for uploads in the languages the help page lists, and you review them in Studio: sign in, select Subtitles from the left menu, click the video, then under Subtitles click More next to the subtitles you want to edit. From there you can “review automatic captions and edit or remove any parts that haven’t been properly transcribed”.
Google is specific about what these are. The page says automatic captions “are generated by machine learning algorithms, so the quality of the captions may vary”, that they “might misrepresent the spoken content due to mispronunciations, accents, dialects, or background noise”, and that creators should always review them.
It also lists the reasons they might not appear at all: the captions are not ready yet because of complex audio, automatic captions do not support the language in the video, the video is too long, the sound quality is poor or YouTube does not recognise the speech, there is a long silence at the start, or several speakers overlap or several languages are used at once.
One more thing worth knowing: you can take the caption text away with you. In the caption editing view, click Options and select Download subtitles. If the track is an automatic one and you want to change it, Google’s steps are to select Duplicate and edit, then Continue, make your changes, and click Publish.
Four reasons to make your own YouTube transcript file
Automatic captions live on the video. A transcript file lives wherever you need it, and that is the difference that usually drives this job.
The first is corrections. If you are going to fix names, product terms and spellings anyway, it is often quicker to start from a clean transcript of the audio and upload a finished SRT than to edit cue by cue in the browser. The second is a written version of the page: a blog post, a documentation page or a course handout built from what you actually said. The third is show notes, timestamps and chapter markers, which come straight out of an SRT because the timings are already there. The fourth is search, both on your own site and for your own sanity six months later when you need the sentence where you explained the pricing.
| What you want | Where it comes from | The steps |
|---|---|---|
| Captions on the video, no work | YouTube’s automatic captions | Nothing, as long as the conditions above are met |
| A copy of the existing caption text | YouTube Studio | Subtitles, pick the video, open the caption editor, Options, Download subtitles |
| Plain text for a post, notes or search | The MP4, transcribed yourself | Content, Menu, Download, then transcribe the MP4 and export TXT |
| Corrected captions on the video | Your own SRT, uploaded back | Subtitles, ADD LANGUAGE, ADD, Upload file, With timing |
Transcribe the MP4 in the browser instead of uploading it
An unlisted video, a client walkthrough, a recording you have not published yet: none of those are things you want to hand to a transcription service. Our transcriber runs OpenAI’s open-source Whisper model on your own graphics card through WebGPU, so the browser reads the MP4 from disk and nothing is uploaded. No account, no length limit.
- Open the site in desktop Chrome or Edge. Firefox, Safari and phones are not supported yet.
- Drop the MP4 on the page. The video track is ignored and only the audio is read, so the file is simply bigger to load than an audio file would be.
- The first time, the browser downloads the speech model, about 200 MB, and caches it.
- Wait. In our test, 71 seconds of audio took 41 to 50 seconds on a desktop with a graphics card, including the model load. A thin laptop takes roughly as long as the recording itself, so a 20-minute video takes about 20 minutes.
- Export TXT for the written version, or SRT for captions.
Two limits to know before you start. The tool is English only for now, and it uses the base version of the model, chosen so it runs on ordinary hardware rather than for top accuracy. Read the transcript through before you publish anything from it. Names, product jargon and accents are where it slips, which is exactly the vocabulary a channel tends to be full of.
Upload the corrected SRT back to the video
Once you have fixed the spellings, put the file back on YouTube. Google’s steps: sign in to YouTube Studio, select Subtitles from the left menu, select the video, click ADD LANGUAGE and choose your language, then under Subtitles click ADD, select Upload file, choose With timing or Without timing, click Continue, choose your file, and select Save.
Pick With timing for an SRT, because the timings are in the file. Without timing is for a plain transcript that YouTube has to line up with the audio itself.
YouTube lists SubRip (.srt) among the basic caption formats it recommends to anyone new to caption files, and WebVTT (.vtt) among the advanced ones. SRT is the safe default. If you have not handled one before, how to make an SRT file covers the format and the small mistakes that make players reject it, and SRT vs VTT vs plain text explains which export to pick for which job.
Our SRT export builds cues at a maximum of 42 characters per line, two lines per cue, one to six seconds each, with an 80 ms gap, which keeps them readable at the bottom of a video.
If you have a video of your own to do, drop it on the transcriber and see what comes back before you decide how much editing it needs.
Frequently asked questions
Can I download someone else’s video to transcribe it? No. YouTube’s help page for downloading uploads says “You can’t download other user’s YouTube videos”, and the Terms of Service prohibit downloading any part of the Service or its Content except as expressly authorized by the Service or with prior written permission. The Studio download applies to your own uploads.
Why is there no Download option on my video? Google lists four reasons: the video has been removed, it has copyright or Community Guidelines strikes, it uses preapproved audio tracks, or you have already downloaded it 5 times in the past 24 hours. The limit resets after 24 hours.
Can I just download the automatic captions instead? Yes. In the caption editor, click Options and select Download subtitles. That is the fastest route if the automatic track is close enough for what you need. To change an automatic track rather than replace it, select Duplicate and edit, then Continue, and Publish when you are done.
Will my transcript say who is speaking? No. The transcript comes out as one stream of text with no speaker labels, which is fine for a single-presenter video and awkward for an interview. For an interview, add the names by hand while you are correcting the spellings.
Sources, checked 15 September 2026
- YouTube Help, “Download YouTube videos that you’ve uploaded”: https://support.google.com/youtube/answer/56100
- YouTube Help, “Use automatic captioning”: https://support.google.com/youtube/answer/6373554
- YouTube Help, “Add subtitles & captions”: https://support.google.com/youtube/answer/2734796
- YouTube Help, “Edit or remove captions”: https://support.google.com/youtube/answer/2734705
- YouTube Help, “Supported subtitle and closed caption files”: https://support.google.com/youtube/answer/2734698
- YouTube Terms of Service, Permissions and Restrictions: https://www.youtube.com/t/terms
- Our own numbers (speed, model size, cue limits) come from our test of the transcriber on this site