Live Conversation Capture
Every conversation becomes a searchable knowledge asset.
Records, transcribes, and diarises live encounters — customer calls, research interviews, advisory sessions, focus groups — and lands the result as a searchable, attributed corpus item. Consent handling and PII redaction are built in, so teams stop relying on memory or hand-typed notes.
Shape
Operational dimensions
Person and system work side-by-side.
Fires when a user asks.
Light state; replaceable any time.
Consumes external data; does not write back.
Inputs
- Live audio/video stream from capture client or call-platform integration
- Participant identifiers and consent state
- Study / topic / project metadata for corpus tagging
- PII redaction and diarisation config
Outputs
- Speaker-attributed transcript with timestamps
- Recording artefact with source reference
- Structured corpus item tagged by participant, topic, and study
- Consent and redaction audit trail
Mechanism
Captures a live qualitative encounter (interview, call, consult, focus group), records it, transcribes it, and lands the transcript plus metadata (participants, topic, study, consent) as a structured corpus item.
Why this is a primitive
Cannot be decomposed — capture-record-transcribe-attach-metadata is one operation. Strip the transcription and you have a recorder; strip the capture client and you have a transcription service; the corpus-landing-with-metadata is what makes the conversation a durable knowledge object. Does NOT synthesise themes (make-sense-of-it), does NOT teach the participant (learn-it).
Where it shows up
Related primitives
Tags
See where it fits.
Primitives are configured into named solution shapes for each client’s domain. The fastest next step is a conversation about which shape fits your problem.
Start a conversation