YouTube transcript + AI summary

Paste a YouTube link to get the full transcript with timestamps, a short summary, chapters and key points. Copy it, or download TXT and SRT.

  • No extension
  • No YouTube account
  • Shorts and youtu.be links
  • 16 languages

Sample result

Sample result

Inside the mind of a master procrastinator | Tim Urban | TED

by TED

Length
14:03
Words
2,277
Chapters
8
Captions
English

AI summary

Tim Urban explains procrastination with three characters: a Rational Decision-Maker, an Instant Gratification Monkey that grabs the wheel, and a Panic Monster that only wakes up near a deadline. Deadlines eventually force the work, but procrastination with no deadline (health, relationships, starting a career) never triggers the Panic Monster, and that is the kind that quietly causes long-term regret.

Key points
  • The Instant Gratification Monkey only cares about what is easy and fun right now.
  • The Panic Monster wakes up when a deadline gets close, which is why last-minute work gets done.
  • Procrastination with no deadline is the dangerous kind: nothing ever forces you to start.
  • Everyone is procrastinating on something; the Life Calendar shows how few weeks there are.
Chapters
  1. The 90-page senior thesis

    Two all-nighters and a paper handed in at the last second.

  2. Inside a procrastinator's brain

    The Rational Decision-Maker and the Instant Gratification Monkey.

  3. The Dark Playground

    Easy, fun time that is full of guilt and dread.

  4. The Panic Monster

    Deadlines and public embarrassment scare the Monkey away.

  5. Procrastinating on this TED Talk

    How the talk itself got written.

  6. Thousands of emails

    Readers describe what procrastination has cost them.

  7. When there is no deadline

    Long-term procrastination that the Panic Monster never stops.

  8. The Life Calendar

    One box per week of a 90-year life.

Written by gpt-6-luna from the transcript. Check the transcript for anything important.

Transcript

  1. So in college,
  2. I was a government major,
  3. which means I had to write a lot of papers.
  4. Now, when a normal student writes a paper,
  5. they might spread the work out a little like this.
  6. So, you know --
  7. (Laughter)
  8. you get started maybe a little slowly,
  9. but you get enough done in the first week
  10. that, with some heavier days later on,
  11. everything gets done, things stay civil.
  12. (Laughter)
  13. And I would want to do that like that.
  14. That would be the plan.
  15. I would have it all ready to go,
  16. but then, actually, the paper would come along,
  17. and then I would kind of do this.
  18. (Laughter)
  19. And that would happen every single paper.
  20. But then came my 90-page senior thesis,
  21. a paper you're supposed to spend a year on.
  22. And I knew for a paper like that, my normal work flow was not an option.
  23. It was way too big a project.
  24. So I planned things out,
  25. and I decided I kind of had to go something like this.
  26. This is how the year would go.
  27. So I'd start off light,
  28. and I'd bump it up in the middle months,
  29. and then at the end, I would kick it up into high gear
  30. just like a little staircase.
  31. How hard could it be to walk up the stairs?
  32. No big deal, right?
  33. But then, the funniest thing happened.
  34. Those first few months?
  35. They came and went,
  36. and I couldn't quite do stuff.
  37. So we had an awesome new revised plan.
  38. (Laughter)
  39. And then --
  40. (Laughter)
  41. But then those middle months actually went by,
  42. and I didn't really write words,
  43. and so we were here.
  44. And then two months turned into one month,
  45. which turned into two weeks.
  46. And one day I woke up
  47. with three days until the deadline,
  48. still not having written a word,
  49. and so I did the only thing I could:
  50. I wrote 90 pages over 72 hours,
  51. pulling not one but two all-nighters --
  52. humans are not supposed to pull two all-nighters --
  53. sprinted across campus,
  54. dove in slow motion,
  55. and got it in just at the deadline.
  56. I thought that was the end of everything.
  57. But a week later I get a call,
  58. and it's the school.
  59. And they say, "Is this Tim Urban?"
  60. And I say, "Yeah."

Showing the first 60 lines. A real run returns the whole transcript.

What you get

Everything in one run, from one link.

Full transcript
Every caption line with its start time. Search it, click a time to play the video there, copy it, or download TXT and SRT.
AI summary
Two to four sentences on what the video says, so you know in seconds whether it is worth the next 40 minutes.
Chapters
Sections with start times, ready to paste into a YouTube description. The first one always starts at 0:00.
Key points
The concrete takeaways: facts, numbers and steps, as a list you can copy as Markdown.

Compared with the usual options

FeatureThis toolYouTube "Show transcript"Browser extensions
Full transcript with timestamps✓ Yes✓ Yes✓ Yes
AI summary, chapters and key points✓ Yes—Some, paid
Copy-ready YouTube chapters (0:00 …)✓ Yes—Rarely
Download TXT and SRT✓ Yes—Some
16 transcript languages✓ YesTranslated captions onlyVaries
Works on phones, nothing to install✓ Yes✓ Yes—
Same data as an API✓ Yes——

Who uses it

  • YouTube creatorsChapters, descriptions and subtitles for every upload, and a draft blog post or thread from each video.
  • Students and researchersNotes and exact quotes with timestamps from lectures, talks and interviews.
  • Marketers and analystsRead competitor videos and webinars in minutes, and search them for products, claims and numbers.

Get YouTube transcripts with an API

The captions come from a SandBase API. Same data, in your own code or Agent.

Send a video ID and a language code; get SRT, TXT, json3 or XML back. Leave the language empty to list the caption tracks a video has.

curl -X POST https://api.sandbase.ai/v1/api/youtube/web-v2/video-captions \
  -H "Authorization: Bearer $SANDBASE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"video_id": "arj7oStGLkU", "language_code": "en", "format": "srt"}'

Frequently asked questions

How do I get the transcript of a YouTube video?
Paste the video link into the box at the top and press Get transcript. You get the full transcript with timestamps, an AI summary, chapters and key points. Copy the text, or download it as TXT or SRT.
Does it work for videos without subtitles?
It works for any video that has captions on YouTube, including the automatic captions YouTube makes for most videos. If a video has no captions at all, because the owner turned them off or YouTube has not made them yet, there is nothing to read.
Can I get the transcript in another language?
Yes. Pick one of 16 languages next to the input. Video language returns the original captions; any other language returns YouTube's translated captions.
How is the AI summary made?
The timed transcript goes to one language model call that writes a short summary, splits the video into chapters with start times and lists the key points. Chapter times play the video from that moment. Check the transcript for anything important.
Can I use the chapters in my YouTube description?
Yes. Press Copy as YouTube chapters to get lines like "0:00 Intro" to paste into your video description. The first chapter always starts at 0:00, as YouTube requires.
Does it work with Shorts, live streams and youtu.be links?
Yes. Watch links, youtu.be short links, Shorts, embeds and past live streams all work, as long as the video is public or unlisted and has captions.
Is there an API for YouTube transcripts?
Yes. Call POST /v1/api/youtube/web-v2/video-captions on SandBase with a video_id, a language_code and a format (srt, txt, json3 or xml). API calls are billed per call to your SandBase balance.

More tools

All tools →

SandBase APIs for this job

The same kind of data in your own code or Agent: one API key, billed per call.

YouTube APIs →DataForSEO APIs →TikTok APIs →All 2,000+ APIs →