RONIN
Meeting Copilot

A local-first meeting copilot for macOS. Real-time transcription, AI-powered suggestions, and post-meeting summaries—audio never leaves your Mac.

macOS 15+ Apple Silicon 128 Tests MIT License

Meeting tools record everything and send it to the cloud. Your conversations, your negotiation strategies, your half-formed ideas—all uploaded to servers you don't control. RONIN takes the opposite approach. Audio stays on your device. Transcription runs on your GPU. The cloud is optional, never required.

Instead of a passive recording, RONIN is an active copilot. It listens in real-time and surfaces suggested responses, follow-up questions, risk flags, and relevant facts from your prep notes—while the conversation is still happening. After the meeting, it distills everything into an executive summary, key decisions, action items, and open questions.

Live Transcription

MLX Whisper running natively on Apple Silicon via Metal GPU acceleration. No cloud API calls, no latency.

Suggested Responses

2–3 tone-varied replies (direct, diplomatic, curious) generated in real-time as the conversation unfolds.

Risk Flags

Alerts when discussion conflicts with your constraints or goals. Surfaces relevant facts from prep notes on cue.

Post-Meeting Summary

Executive summary, key decisions, action items, and open questions—generated the moment the meeting ends.

Four LLM Providers

Apple Intelligence (fully on-device, macOS 26+), LM Studio (local), OpenAI, or Anthropic. Your choice, your control.

Zero Conflicts

AVCaptureSession mic capture works alongside Teams, Zoom, and WhatsApp without aggregate device hacks.

Two paths, one principle. The backend path routes audio through a Python FastAPI server running MLX Whisper for transcription and a pluggable LLM client for copilot intelligence. The Apple Intelligence path keeps everything on-device—transcription via the backend in transcription-only mode, then Swift Foundation Models generate responses locally with chunked map-reduce for long meetings.

SwiftUI Mac App
  ├── MeetingPrep → LiveCopilot → PostMeeting
  ├── AudioCaptureService (AVCaptureSession)
  ├── NativeCopilotService (Apple Intelligence path)
  └── WebSocket connection
            │
Python Backend (FastAPI)
  ├── MLX Whisper (transcription)
  ├── LLM Client (pluggable)
  │     ├── LM Studio (local)
  │     ├── OpenAI (cloud)
  │     ├── Anthropic (cloud)
  │     └── none (transcription only)
  └── Meeting State Manager
128
Tests Passing
5
E2E Tests
4
LLM Providers
0
Cloud Required
M1+
Apple Silicon
MIT
License
Swift 5.10 SwiftUI Python 3.13+ FastAPI MLX Whisper Apple Intelligence Foundation Models WebSocket XcodeGen macOS 15+