Speaker Diarization Multi Speaker Attribution

Patient arrived 5 minutes early and began describing symptoms to the triage nurse while recording was already active. First 31 seconds of audio capture the patient saying 'I've been having chest tightness for two weeks, worse wit…

Agent assigns PATIENT to speaker_A based on first-person symptom self-report framing, subjective complaint structure, and duration/severity quantification. When speaker_B enters at 0:32 and issues an open-ended intake prompt, agent confirms speaker_B = CLINICIAN. Chest tightness with exertion appea…

The system incorrectly assigned CLINICIAN to the patient at utterance 1 (patient spoke first). Over utterances 2–13, the 'patient' cluster accumulates: first-person self-report ('my pain is a 7'), lay hedging ('I'm not sure but..…

Agent fires a global retroactive role flip: reassigns all 12 prior utterances from their current (wrong) labels to corrected labels. Re-processes every SOAP field that drew from inverted utterances, moving content from Assessment back into HPI and vice versa as appropriate. Logs each reassigned utt…

The patient, a registered nurse, describes her chief complaint using precise clinical language from utterance 1: 'I've had intermittent paroxysmal supraventricular tachycardia episodes — heart rate around 180 — no syncope, no dia…

Agent assigns PATIENT to speaker_A based on: (a) first-person ownership framing ('I've had'), (b) self-report symptom structure with onset and associated-symptom enumeration, (c) turn-taking position as respondent to an implicit intake prompt, not as an issuer of questions or directives. Agent does…