YouTube Captions Memory

YouTube Captions Memory lets Hana ingest caption tracks from YouTube videos and use that text as searchable memory context.

What You Can Do

• Add a YouTube video URL as a memory source.
• Choose the caption language track to ingest.
• Add optional use-case instructions for retrieval behavior.
• Sync, retry, abort, delete, and search ingested YouTube memory batches.

Supported Input

• youtube.com and youtu.be video URLs.
• Valid caption tracks for the selected language.

How to Ingest

1. Open Memory in Hana Dashboard.
2. Select the YouTube Memory tab.
3. Click Add YouTube Captions Memory.
4. Enter:
   • YouTube link
   • Caption Language (optional advanced field)
   • Use case instructions (optional)
5. Confirm submission and monitor ingestion status until INGESTED.

Operational Actions

After creation, the row actions support standard memory lifecycle controls:

• Sync to refresh captions.
• Retry if ingestion failed.
• Abort while ingestion is running.
• Delete completed/errored batches.

Tips

• Prefer videos with reliable caption tracks for best results.
• Use explicit language selection when videos provide multiple caption options.
• Add concise use-case instructions when you want Hana to focus on a specific extraction style (for example, “summarize tactical steps only”).

Example Invocations

@Hana summarize the key takeaways from the YouTube video memory we added yesterday.
@Hana extract implementation steps from the YouTube tutorial memory in bullet points.

See also

• Common Memory Module Information
• Google Drive Memory
• Auto Resync Memory