Skip to content
Try it free

Download a YouTube Transcript

Paste a link, pick a format, download. SRT, VTT, Markdown and JSON keep the timestamps.

1. Pick a format

For video editors. Same words and timing in every format.

example.srt
100:00:03,200 --> 00:00:05,800Every format starts from the same transcript.200:00:05,800 --> 00:00:08,900The timing and the words stay identical,300:00:08,900 --> 00:00:11,400only the wrapping around them changes.400:00:12,600 --> 00:00:15,100SRT numbers the cues and uses commas.500:00:15,100 --> 00:00:18,000VTT adds a header and uses dots.600:00:18,000 --> 00:00:20,700Plain text drops the times entirely.

2. Paste the link

Free to try, no sign-up. You can change the format on the result page.

Open a recorded example

How it works

  1. Pick a format

    SRT for editors, VTT for players, then TXT, Markdown or JSON.

  2. Paste the link

    A watch URL, a youtu.be link or a Short. The file is built for that one video.

  3. Download or print

    Use Export on the result page to save the file, or print it to a PDF.

One transcript, five wrappings

Sample

For video editors

sample.srt
100:00:03,200 --> 00:00:05,800Every format starts from the same transcript.200:00:05,800 --> 00:00:08,900The timing and the words stay identical,300:00:08,900 --> 00:00:11,400only the wrapping around them changes.400:00:12,600 --> 00:00:15,100SRT numbers the cues and uses commas.500:00:15,100 --> 00:00:18,000VTT adds a header and uses dots.600:00:18,000 --> 00:00:20,700Plain text drops the times entirely.

The same six lines in each format. Pick one above to see it with your link.

Which format do you need

The words are the same in every format. Only the wrapping changes, and TXT is the one without times.

FormatUse it forLooks like
SRTVideo editors, YouTube StudioNumbered cues, times with a comma
VTTWeb video playersA WEBVTT header, times with a dot
TXTReading and pastingParagraphs, no times
MarkdownNotes appsA timestamp above each paragraph
JSONCode and agentsSegments with start, end and text
PDFSharing and printingThe whole result, from the print dialog

With timestamps, no extension

To copy the text with timestamps, open Copy on the result page and choose Transcript with timestamps. To save it, SRT, VTT, Markdown and JSON keep the times. For a PDF, choose Print / Save as PDF in the Export menu.

Nothing to install, and no Show transcript button needed. When a video has no captions, Scribiz listens to the audio and writes the transcript. To read it in the browser first, open the YouTube transcript page.

Markdown suits a notes app, because each paragraph starts with its time. JSON holds the segments with start, end and text, which is the one to feed to code.

Transcripts, not the video

This page makes text files. It does not give you the video or its audio. For subtitle files made for a player or an editor, use the YouTube subtitle downloader.

For your own video, YouTube Studio also lets you download the captions already on it. A transcript belongs to the video's creator, so use it for study, notes and accessibility, and ask before you republish it. Privacy and data says what Scribiz keeps.

Limits

One video per run on the web. Public videos only. Without an account: videos up to 15 minutes, 10 Listen or Watch minutes a day and 30 caption lookups a day. A free account takes videos up to 2 hours and 30 minutes a month. Pro takes 6 hours and 600 minutes a month, and is not on sale yet. A video that already has captions uses a tenth of a minute per minute of video.

See pricing · How minutes are counted · Paid plans are not on sale yet.

Questions

Paste the link above, pick a format and press the button. The result page exports SRT, VTT, TXT, Markdown, Context and JSON.

Run the link, open Export on the result page and choose Print / Save as PDF. The PDF is the whole result laid out, and there is no separate PDF button.

Both are timed text. SRT numbers each cue and writes times as 00:00:03,200, while VTT starts with a WEBVTT line and uses a dot. The YouTube subtitle downloader goes into when to use each.

You get the language that is spoken in the video. Translation is not built yet, and machine-translated caption tracks are skipped.

Not on the web, where each run is one video. A script can call the API once for each link.

Captions carry the creator's own timing, which can be off if the video was edited afterwards. Run the video in Listen and the timing comes from the audio.