About Archivarix Tube Search
Archivarix Tube Search is an independent research and archival tool that provides a search interface for publicly archived YouTube video metadata. Our mission is to support digital preservation, academic research, and the public interest by making historically archived web data accessible and searchable.
How It Works
The internet is constantly changing. Web pages, including YouTube video pages, are routinely captured and preserved by public web archiving initiatives such as the Internet Archive (Wayback Machine) and Common Crawl. When a YouTube video becomes unavailable for any reason, the metadata that was previously captured by these archives — including titles, descriptions, upload dates, and text-based subtitles — may still be accessible through their public APIs and datasets. Archivarix Tube Search aggregates and indexes this publicly archived data to make it searchable.
What You Can Do
- Search by YouTube channel URL, @handle, or Channel ID to browse indexed video metadata
- Discover archived metadata for videos that are no longer available on YouTube
- Access text-based subtitles preserved in public web archives
- Check whether video files have been preserved by the Wayback Machine
- Full-text search across indexed video titles, descriptions, and subtitle text
- Generate AI subtitles via speech recognition for videos that have no archived captions — including deleted videos with preserved audio
- Generate AI summaries (TL;DR, key points with timestamps, detailed overview, topic tags) from any transcript — works for deleted videos too
- Generate stenograms — full structured dialogue transcripts with speaker labels, suitable for interviews, podcasts, lectures and panels — also available for deleted videos
AI-Powered Tools
Beyond raw archive search, Tube Search can extract additional value from what is preserved:
- AI Subtitles. When a video has no archived captions but its audio is reachable (live or via preserved files), automatic speech recognition transcribes the speech. The resulting subtitles are stored alongside archived ones and become fully searchable.
- AI Summaries. A structured summary — TL;DR, key points with timestamps, detailed overview, topic tags — generated from any transcript. Because the summary is built from the transcript, it works equally well for videos that have already been deleted from YouTube, as long as a transcript exists in our index.
- Stenograms. A full text dialogue rebuilt from the transcript with speaker labels, formatted as a clean reading transcript. Useful for interviews, podcasts, lectures and panels. Like summaries, stenograms work for deleted videos that still have an archived transcript.
Data Sources
All data presented by this Service is derived from publicly available sources: the Wayback Machine CDX API (Internet Archive), the Common Crawl open dataset, and the YouTube Metadata 2019 research dataset. We do not scrape, crawl, or otherwise access YouTube directly for the purpose of data collection. Thumbnails and subtitle text are retrieved from archived snapshots stored by third-party archives. All videos link directly to their original YouTube page.
Not Affiliated with YouTube
Archivarix Tube Search is not affiliated with, endorsed by, or connected to YouTube, Google LLC, or any of their subsidiaries. YouTube is a registered trademark of Google LLC. This Service is an independent tool that indexes publicly archived data.
Content Removal
If you are a rights holder and believe that metadata displayed on this Service infringes your rights, please contact us using the information below. We maintain a content removal process and will respond promptly to valid requests. See our Terms of Service for details.
Built by Archivarix
This project is developed by the Archivarix team, known for tools that help recover and work with archived web content. Visit archivarix.com to learn more about our other projects.