Fetch YouTube captions with smarter track detection and cleaner exports.

Paste a public YouTube link, detect available caption tracks first, preview the converted transcript, and export a local SRT file. This update improves reliability for manual and auto-generated captions instead of guessing a transcript payload directly.

🔗 YouTube video URL

Tip: public videos with manual or auto-generated captions work best.

✨ Ready! Paste any YouTube URL with captions and click “Fetch captions”.

What was fixed

The fetch logic now requests YouTube's transcript list first, chooses a real caption track, then downloads timed text through multiple public proxy fallbacks. That avoids the earlier "Transcript payload was not found" failure for many videos.

What it supports

Public YouTube links, Shorts URLs, watch URLs, embed URLs, and direct 11-character video IDs.

How it works now

Loads available track metadata, prioritizes English/manual captions when present, then falls back to auto-generated tracks.

Why contributors matter

Future improvements can add transcript language pickers, metadata display, and stronger region-restriction handling.

Project notes

Duck Cloud deployment-friendly and open-source ready.

Designed to work well on simple deployments while exposing cleaner route names like /youtube.
Captions are fetched client-side and downloaded files stay local to the visitor’s browser session.
Open-source contributors are welcome to improve YouTube reliability, UI polish, and documentation.

Fetch YouTube captions with smarter track detection and cleaner exports.

📝 SRT preview

Duck Cloud deployment-friendly and open-source ready.