ClinicalCoach External Module

Description

• Provides a React-based interface for recording audio sessions between a Coach and a Learner.
• Leverages AI (via the SecureChatAI module) to transcribe and analyze each session.
• Consolidates session reflections and final summaries into a structured JSON output.

Key Features

• Audio Capture & Transcription: Integrates Whisper for quick audio-to-text conversion.
• AI Summaries & Reflections: Uses multiple system prompts to generate insights on clinical reasoning, reflective thinking, and final conclusions.
• Logging & Debugging: Includes emDebug calls at key points for transparent, trackable processing.
• Extendable & Modular: Designed to slot into a broader coaching or clinical reasoning ecosystem.

Workflow Highlights

1. Audio Input: Upload or record audio from the user interface.
2. Transcription: The EM calls Whisper API to convert audio to text.
3. AI Analysis: Summaries, reflections, and final outputs are generated through a sequence of LLM prompts.
4. Structured Results: Returns a standardized JSON object containing all AI-produced content, including error handling.

Dependencies

• SecureChatAI module for calling Whisper and GPT-like endpoints.
• React Front End built and bundled into MVP/dist/assets.