Decision record
Separates decisions, open questions, objections, and commitments instead of flattening the meeting into a generic summary.
Evidence-grounded meeting review
Turn a meeting recording into decisions and accountable actions.
VideoLens reviews meeting video and audio for decisions, objections, commitments, and follow-ups, then anchors the report to the timestamps where those moments occurred.
Transcript + frame vision + OCR
Timestamp citations
~1,500 supported platforms
Markdown, PDF, and JSON
The direct answer
What VideoLens does
Meeting summaries are most useful when a teammate can verify what was decided and why. VideoLens combines the transcript with visible presentation context and produces evidence-linked findings instead of a memory-based recap.
Use Meeting mode for a structured review, or ask a narrower question about a customer objection, pricing discussion, unresolved decision, or owner commitment.
Why it helps
From linear video to reviewable evidence
Timestamp accountability
Citations let owners verify the precise discussion behind an action item.
Visual meeting context
Sampled frames can capture slides, shared screens, and other visible context that a transcript omits.
Workflow
How the analysis works
Every stage is explicit and cached so the source can be checked and the analysis can be reused.
Provide the permitted meeting recording as a local file or accessible URL.
Choose Meeting mode and state the decisions or themes you care about.
VideoLens merges transcription, sampled visual context, and visible text into the timeline.
Confirm owners and decisions, ask follow-ups, and export the final record.
Output
What the report gives you
- Meeting summary and key themes
- Decisions, objections, and unresolved questions
- Commitments and follow-up actions
- Timestamp evidence for important claims
Important limitation: Speaker attribution depends on available diarization and recording quality and may be imperfect. Confirm names, owners, dates, and consequential commitments before distributing the report.
FAQ
Common questions
Can VideoLens identify different speakers?
It can use diarized transcription when available, but speaker attribution quality depends on the audio and transcription output. Human verification is recommended.
Can I ask about one topic instead of summarizing the whole meeting?
Yes. Use a focused prompt, such as asking for every pricing objection or the evidence behind a specific decision.
Are meeting recordings stored by VideoLens?
The open-source workflow runs locally. The hosted app uses ephemeral session processing and your OpenAI key; VideoLens does not build a retained meeting dataset.
Related guides
Explore more VideoLens workflows
Turn the next video into evidence.
Try the hosted app, self-host the MIT-licensed core, or connect VideoLens to an MCP client.