Transcripts your AI can quote back.

The pain

• —30 customer interviews in your Notion. The AI summarized each one — but the cross-interview patterns are stuck in your head.
• —Meeting transcripts pile up. "What did Sarah say about pricing in Q2?" → 20 minutes of scrolling.
• —Action items live in one place, decisions in another, raw transcripts in a third — none of them deploy to Claude when you ask for a synthesis.
• —Six months later, you can't trace why you pivoted. The conversation that turned the team is buried.

What you do in Memory.Wiki

1. 1
   Capture the transcript

   Paste it from Otter / Fireflies / Granola, or import from a Notion page that holds it. Memory.Wiki normalizes the speakers and saves a permanent URL.
2. 2
   Tag with intent

   Mark each doc as note / decision / question. Doc intent is auto-classified after the next concept refresh; you can override with the chip in the editor's LIVE bar.
3. 3
   Bundle by project

   Group all customer interviews for one project into a Bundle. Run Discoveries — the AI surfaces tensions ("Customer A wants more X; Customer B wants less X") and gaps automatically.
4. 4
   Recall across interviews

   Open the Hub Assistant and ask "What did customers say about pricing?" Hub recall hits every interview that mentions pricing and quotes the speaker + line directly.

What you get back

• ✓"Why did we pivot?" → the AI cites the three interviews that pushed the decision.
• ✓Action items get an intent=decision tag; sidebar filter shows every decision the team made this quarter.
• ✓Synthesis docs compiled from the bundle remember their source — Recompile after you add a new interview, get an updated synthesis.
• ✓Send the hub URL to a new hire and they're up to speed on the customer landscape in an afternoon.

Worked example

Try it with what's on your desk right now.

No signup. Drop in your first doc and the URL is yours.

Other shapes the URL takes