Recording views and visual evidence
Recap, Transcript, and Artifacts arrive from core processing. Deep Dive is the optional fourth view for video, with clear progress while visual evidence is created.
- Updated
- 30 August 2026
- Read
- 5 min
- For
- Everyone
What each view answers.
They share one source, so anything you find in one can be opened in the others.
The short version: a summary with the decisions, action items and owners, and topics pulled from what was actually said. Each key outcome seeks the player to the moment behind it, so the recap is a map of the evidence rather than a replacement for it.
Everything said, with speakers separated and timestamps throughout. Read it as a stream or grouped by speaker. Clicking any line jumps the player there.
The optional visual record: what was on screen at each moment, scene by scene, with verified keyframes clickable. It appears with progress after you request visual evidence and becomes usable when that generation is ready. Audio-only recordings do not offer it.
Typed transcript findings are available before visual analysis, carrying quote, speaker, and timestamp. Visual evidence can corroborate those same rows or add genuinely new visual-only findings without replacing IDs or review state.
Getting to the moment.
- Everything seeks the player
Transcript lines, key outcomes, findings, scene thumbnails, and Q&A citations all jump the recording to their moment. You rarely need to scrub.
- Chapters
After visual evidence is ready, scenes group into chapters beside the player, so a long session can be skimmed by what was on screen rather than by time.
- Speakers
The speaker panel shows who spoke and for how long. Names come from the meeting where the assistant captured it; upload speakers can be tagged by hand.
- Rename it
The title is editable from the recording page, and suggested titles are offered based on what the session contained.
Questions across the whole session go through Evidence Q&A.
Recording views, answered.
- Audio-only files do not offer Deep Dive. For video, the tab appears with progress after you request Analyze visual evidence and becomes usable after the generation is verified. Until then the transcript-first tabs remain available without an empty visual tab.
- Open the moment behind it. Every recap line seeks the player, so you can hear exactly what it was drawn from and judge for yourself - that is the point of a cited record.
- Not today. You can correct who spoke by tagging speakers, which updates the labels used in the transcript and in future citations.
- The recording had no participant names to draw from, which is normal for uploads. Tag the speaker once on the recording page and the name is used from then on.
- The recap summarises the whole session. A Q&A answer responds to your specific question and cites the passages it used. Both link back to the same moments.