The result: Consultation notes are generated automatically from voice: after-hours charting turns into a review of a draft that is already written.

The problem

Clinicians lose hours to note-taking after every consultation. Typing up what was said, pulling out the clinical details and writing a summary happens after the patient has gone, often at the end of a long day.

What we built

  • A voice-first platform built on Django.
  • AssemblyAI speech-to-text to capture the consultation.
  • Claude, Groq and Gemini for clinical entity extraction, multi-turn chat and automatically generated summaries.

Stack

Django · AssemblyAI · Claude · Groq · Gemini

What this means for your business

The same pieces, speech-to-text, structured extraction and an assistant that works from a defined set of records, power intake lines, call summaries and voice agents in other industries.

Related: AI voice agent development · AI for clinics · All case studies

Client names are kept private. Kortex Labs is founder-led by Saif Ur Rehman, who built this project.