frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

I built a tool to generate Cornell notes from YouTube videos

3•cristyg0101•4h ago•1 comments

Ask HN: Anyone set up ways to easily obtain and read transcripts from Ted, YT?

2•MollyRealized•9h ago•4 comments

Where are people finding GPU capacity?

3•adilhafeez•15h ago•2 comments

Continue coding agent is dead. Alternatives?

3•roscas•1d ago•3 comments

Coding Agents killed my identity. How do you feel?

20•bah9•1d ago•33 comments

Windmill.dev is not "fully open-sourced (AGPLv3)"

9•alpine-rockies•1d ago•2 comments

Ask HN: What's the current best research/understanding of healthy weight loss?

2•simonebrunozzi•13h ago•4 comments

Ask HN: At what point does "clean architecture" become technical debt?

4•kiops•19h ago•4 comments

Ask HN: OpenCode no longer including DeepSeek?

8•dv35z•1d ago•6 comments

Ask HN: Why does the US government search through people's phones at the border?

13•roschdal•1d ago•12 comments

Robotics Software

3•erardaiden•1d ago•0 comments

Ask HN: Why is HN missing a "go to top" link on long posts?

4•KellyCriterion•1d ago•5 comments

Ask HN: How to Relicense an Old Project

3•gradschool•1d ago•1 comments

SearXNG running on OpenBSD on a ThinkPad t420

2•open-search•1d ago•0 comments

You've reached the end!

Open in hackernews

Ask HN: Anyone set up ways to easily obtain and read transcripts from Ted, YT?

2•MollyRealized•9h ago
I really absorb text a great deal better than listening to audio or watching video. I've heard others say the same and attribute to ADHD, so perhaps that's the case for me.

TED very often has a transcript on their page, but it involves some clicking and then some copy-pasting, and the text result is often broken mid-clause after about 10-15 characters.

I'm wondering if there's any useful userscripts, userstyles, or other methods that handle this. (I think TED often sues people who try to make transcripts easily available on a separate site. I may be mistaken, and TED, please don't sue me, it's just a general feeling.)

The same request would apply to YouTube and/or other popular video sites, but I run into this issue most often with TED.

My thanks in advance to anyone who can help me on this!

Comments

toomuchtodo•9h ago
https://github.com/yt-dlp/yt-dlp

https://news.ycombinator.com/item?id=44381640

MollyRealized•8h ago
yt-dlp --skip-download --write-sub --write-auto-sub --sub-lang en --sub-format json2 "[1]"

[TedTalk] Extracting URL: [1]

[TedTalk] luvvie_ajayi_jones_how_to_be_a_professional_troublemaker: Downloading webpage

ERROR: the JSON object must be str, bytes or bytearray, not NoneType

exit 1

[1] - full URL is: https://www.ted.com/talks/luvvie_ajayi_jones_how_to_be_a_pro...

sandreas•7h ago
Did you try

  yt-dlp --update-to master
YouTube tries to shut down yt-dlp and sometimes it is good to have the absolute latest code.
loveparade•6h ago
I built my own pipeline for this because in my experience the YouTube auto transcripts are quite bad.

I download the raw audio track with yt-dlp, give it to qwen asr for raw transcription, then DeepSeek flash for transcription cleanup, formatting, corrections with web search enabled, etc. I use openrouter to call the models. It's a few cents per video (mostly the asr model at ~12 cents per hour) and quality is great. You could probably use a local asr model if you want to save money, I just don't bother because it's so cheap already.

That only makes sense for videos that don't have official (not auto-generated) transcripts of course. If there are official transcripts that are good you can just get them directly with yt-dlp.