DEDUPED Hot AI repos, turned into use cases Get the Monday 5

How to transcribe Zoom and Teams meetings with speaker labels with Openwhispr

Let OpenWhispr detect calls, transcribe them with local speaker diarization and save searchable notes.

Short answer: Install OpenWhispr; then connect Google, Microsoft or Apple Calendar so calls are detected; Join a Zoom, Teams or FaceTime call and let it transcribe with live speaker diarization. All steps.

Who it is forA manager or consultant who wants Granola style meeting notes without sending recordings to a vendor.

The problemMeeting note tools are paid and upload your calls, which some clients and companies will not allow.

The outcomeMeeting transcripts with speaker labels and voice fingerprints, stored as searchable notes.

What you need first

  • macOS, Windows or Linux
  • Optional: local models download for offline use, or cloud provider keys
  • To run from source: Node.js 24+

Steps

  1. Install OpenWhispr
  2. Connect Google, Microsoft or Apple Calendar so calls are detected
  3. Join a Zoom, Teams or FaceTime call and let it transcribe with live speaker diarization
  4. Find the transcript in Notes and use semantic search or AI actions
  5. Import older recordings with audio import

Prompt for your agent

Using OpenWhispr/openwhispr, set up meeting transcription with local speaker diarization for my <calendar> calls, then after my next meeting summarize the transcript into decisions and action items per speaker.

Caveat. Tell participants they are being transcribed and follow local consent laws. On Intel Macs, live speaker identification and voice fingerprinting are unavailable.

Use this in your AI agent

Ask your agent

Using the Deduped MCP, get the setup steps for OpenWhispr/openwhispr and "Transcribe Zoom and Teams meetings with speaker labels"

Add Deduped to Claude Code once

claude mcp add --transport http deduped https://dedupedai.com/mcp

Cursor, ChatGPT, others

Share: X LinkedIn Hacker News Email Copy link

Third-party repos we don't own. AI-written use cases can be wrong. Disclaimer