Meeting Knowledge Extraction

Captures meeting content and turns it into summaries, notes, and follow-up drafts for later use.

Business impact

• meeting-record creation time — NEC reported a reduction from about 30 minutes to about 5 minutes.
• follow-up turnaround time — Morgan Stanley reported that follow-ups that previously took days now happen within hours.

Data requirements

• meeting recordings (audio) — Used in Morgan Stanley’s Debrief to produce notes and draft follow-ups.
• meeting discussion content (text) — Used to generate summaries and meeting records.
• internal meeting records (text) — The output becomes a record for later use.

AI methods and techniques

• speech transcription — Used to process meeting recordings into text for later drafting.
• summarization — Used to produce meeting summaries and records from the discussion.
• draft generation — Used to create draft follow-ups for human review.
• context interpretation — Used in meeting support to include contextual nuances in summaries.

AI models and model families

Whisper, GPT-4

Industries

Real-world evidence

3 documented case studies on record.

Companies using this: Honda, Morgan Stanley, NEC.