MeetingScribe

High-quality Teams meeting audio recorder - Launch instantly via Raycast.

A streamlined, production-ready tool for recording Microsoft Teams meetings with professional WASAPI audio quality. Lightweight, fast, and focused exclusively on recording.

Quick Start

Installation

1. Clone and setup:

git clone <repository-url>
cd meetingscribe
python scripts/bootstrap.py

2. Configure environment:

cp .env.example .env
# Edit .env if needed (defaults are optimized for Teams)

3. Verify installation:

python scripts/system_check.py

Usage

Raycast Extension (Recommended):

1. Import extension from raycast-extension/
2. Configure preferences:
   • Python executable path
   • MeetingScribe project directory
   • Optional: Google Gemini API key (for transcription features)
3. Available commands:
   • "Start Recording" - Begin recording Teams meeting
   • "Recent Recordings" - Browse and manage audio files
   • "Recording Status" - Monitor active recording sessions
   • "Transcription Progress" - Track transcription jobs (if enabled)
4. Recordings automatically saved to storage/recordings/

Command Line:

# Record for 30 seconds
python -m cli record 30

# Record for 5 minutes with M4A format
python -m cli record 300 m4a

# Check system status and audio devices
python -m cli status

Project Structure

meetingscribe/
├── src/                          # Core Python application
│   ├── audio/
│   │   ├── recorder.py          # WASAPI audio recording engine
│   │   ├── devices.py           # Audio device detection & selection
│   │   └── __init__.py
│   ├── teams/
│   │   ├── teams_detector.py    # Teams process detection
│   │   ├── integration.py       # Teams integration helpers
│   │   └── __init__.py
│   ├── cli/
│   │   ├── main.py              # CLI entry point & commands
│   │   ├── __main__.py
│   │   └── __init__.py
│   ├── config.py                # Pydantic-based configuration
│   └── transcriber.py           # Audio processing utilities
│
├── raycast-extension/            # Raycast launcher (TypeScript)
│   ├── src/
│   │   ├── record.tsx           # Start recording
│   │   ├── recent.tsx           # Browse recordings
│   │   ├── recording-status.tsx # Monitor recording sessions
│   │   ├── transcription-progress.tsx  # Track transcription
│   │   └── cli.ts               # Python CLI helper
│   ├── package.json
│   └── tsconfig.json
│
├── scripts/                      # Utility scripts
│   ├── bootstrap.py            # Setup & installation
│   └── system_check.py         # System diagnostics
│
├── storage/                      # Runtime data (generated)
│   ├── recordings/             # WAV/M4A audio files
│   ├── logs/                   # Application logs
│   └── transcription/          # Transcription progress
│
├── docs/                         # Documentation
│   ├── USER_GUIDE.md           # Installation & usage
│   ├── DEVELOPMENT.md          # Developer guide
│   ├── ARCHITECTURE.md         # System design
│   └── usage.md
│
├── tests/                        # Test suite
├── tools/                        # Development tools
├── .vscode/                      # VS Code settings
│
├── .env.example                  # Configuration template
├── pyproject.toml               # Python metadata
└── requirements.txt             # Python dependencies

Features

✅ Professional Audio Quality: 48kHz, 16-bit, stereo WASAPI loopback recording ✅ Multiple Formats: WAV (uncompressed) and M4A (compressed) output options ✅ Instant Raycast Launch: Cmd+Space + "Start Recording" quick access ✅ Auto Device Detection: Intelligent WASAPI loopback selection ✅ Real-Time Monitoring: Live recording status and progress tracking ✅ Lightweight: Minimal dependencies, fast startup and low overhead ✅ Teams Optimized: Designed for Microsoft Teams meeting capture ✅ Optional AI Transcription: Integrate Google Gemini for transcription (separate feature)

Documentation

• User Guide: docs/USER_GUIDE.md - Installation and usage
• Developer Guide: docs/DEVELOPMENT.md - Contributing and development
• Architecture: docs/ARCHITECTURE.md - System design

Requirements

• OS: Windows 10/11 (WASAPI loopback)
• Python: 3.9+
• Dependencies: See requirements.txt

Design Philosophy

Core Focus: Exceptional audio capture, nothing else

• ✅ Professional WASAPI loopback recording
• ✅ Teams meeting optimization
• ✅ Configurable duration and formats
• ❌ No transcription (recording only)
• ❌ No speaker detection or analysis
• ❌ No built-in AI processing (optional external transcription)

Why? A focused tool does one thing exceptionally well. Record with confidence, transcribe separately with tools of your choice.

Tech Stack

Backend (Python)

• pyaudiowpatch - WASAPI loopback support
• pydantic - Configuration & data validation
• loguru - Advanced logging
• pywin32 - Windows audio APIs

Frontend (Raycast Extension)

• TypeScript & React
• Raycast API
• Real-time status monitoring

Getting Started

1. Clone repository and run setup:

   python scripts/bootstrap.py

2. Verify your system:

   python scripts/system_check.py

3. Import Raycast extension from raycast-extension/ directory

4. Configure preferences (Python path, project directory, optional API key)

5. Launch recording: Cmd+Space → "Start Recording"

Development

See docs/DEVELOPMENT.md for:

• Development environment setup
• Common commands
• Testing procedures
• Contributing guidelines

Architecture

See docs/ARCHITECTURE.md for detailed system design, component descriptions, and data flow.

License

MIT License - See LICENSE

Support & Feedback

• System Check: python scripts/system_check.py - Diagnose issues
• Documentation: docs/ - Detailed guides
• Issues: Check documentation first, then file issues with system check output