Voice Mode

# LiveKit Service Implementation Summary ## Overview This document summarizes the implementation of LiveKit as a managed service in Voice Mode, created during the night session of 2025-08-12. ## What Was Implemented ### 1. Core Service Files #### Installation (`voice_mode/tools/services/livekit/install.py`) - Platform detection (macOS uses Homebrew, Linux uses official script or direct download) - Binary installation with version management - Configuration file generation (dev mode with devkey/secret) - Service file installation - Auto-enable option for startup #### Uninstallation (`voice_mode/tools/services/livekit/uninstall.py`) - Service stopping and disabling - Binary and configuration removal - Optional complete data cleanup - Homebrew detection and warning #### Helper Functions (`voice_mode/utils/services/livekit_helpers.py`) - `find_livekit_server()` - Locate LiveKit binary - `find_livekit_config()` - Locate configuration file - `check_livekit_health()` - Health check via HTTP - `get_livekit_version()` - Version detection - `is_livekit_installed()` - Installation check ### 2. Service Templates #### systemd (`voice_mode/templates/systemd/voicemode-livekit.service`) - Proper service configuration - Health check on startup - Resource limits (1GB memory, 50% CPU) - Environment variables for dev credentials - Journal logging #### launchd (`voice_mode/templates/launchd/com.voicemode.livekit.plist`) - macOS service configuration - Auto-start at login - Log file paths - Environment setup including PATH ### 3. Configuration Updates #### `voice_mode/config.py` Added LiveKit configuration constants: - `LIVEKIT_PORT` (default: 7880) - `LIVEKIT_URL` (default: ws://localhost:7880) - `LIVEKIT_API_KEY` (default: devkey) - `LIVEKIT_API_SECRET` (default: secret) ### 4. Service Management Integration #### `voice_mode/tools/service.py` Extended to support LiveKit: - Added LiveKit to service name literals - Updated `get_service_config_vars()` for LiveKit - Modified all port selection logic to include LiveKit - Updated docstrings ### 5. CLI Commands #### `voice_mode/cli.py` Added complete LiveKit command set: - `voice-mode livekit install` - Install LiveKit server - `voice-mode livekit uninstall` - Remove LiveKit - `voice-mode livekit start/stop/restart` - Service control - `voice-mode livekit enable/disable` - Startup management - `voice-mode livekit status` - Check service status - `voice-mode livekit logs` - View service logs - `voice-mode livekit update` - Update service files ## Key Design Decisions ### 1. Dev Mode by Default Following Mike's guidance, the implementation uses LiveKit's dev mode with standard credentials (devkey/secret) for easy setup. ### 2. Platform-Specific Installation - **macOS**: Prefers Homebrew if available - **Linux**: Uses official install script, falls back to direct binary download - **Both**: Install to `~/.voicemode/services/livekit/` for consistency ### 3. Service Pattern Consistency LiveKit follows the same patterns as Whisper and Kokoro: - Same directory structure - Same service management commands - Same configuration approach - Same health check pattern ### 4. Configuration File Creates a YAML configuration file with: - Dev mode keys - Port configuration - RTC port range - Logging settings - Room auto-creation ## What's Still Needed ### 1. MCP Tool Registration The `install_livekit` and `livekit_status` tools need to be registered in the MCP server. ### 2. Provider Discovery LiveKit should be added to the provider discovery system so it's automatically detected when running. ### 3. Web Interface The web interface in `docs/web/` needs to be updated to: - Auto-detect local LiveKit server - Configure connection settings - Handle fallback to cloud ### 4. Documentation - Update main README with LiveKit information - Create user guide for LiveKit setup - Update configuration documentation ### 5. Testing - Test installation on both macOS and Linux - Verify service management works correctly - Test web interface connection - Verify TailScale access ## Usage Examples ### Installing LiveKit ```bash # Basic installation voice-mode livekit install # Custom port voice-mode livekit install --port 8880 # Force reinstall voice-mode livekit install --force # Install and enable at startup voice-mode livekit install --auto-enable ``` ### Managing the Service ```bash # Check status voice-mode livekit status # Start/stop voice-mode livekit start voice-mode livekit stop # View logs voice-mode livekit logs --lines 100 # Enable at startup voice-mode livekit enable ``` ### Uninstalling ```bash # Basic uninstall voice-mode livekit uninstall # Remove everything voice-mode livekit uninstall --remove-all-data ``` ## Next Steps 1. Complete MCP tool registration 2. Add provider discovery support 3. Update web interface for LiveKit connection 4. Test the complete workflow 5. Create user documentation 6. Submit PR for review ## Notes This implementation follows the existing patterns in Voice Mode closely, making it feel like a natural extension rather than a bolt-on feature. The dev mode configuration makes it extremely easy to get started, which aligns with Voice Mode's philosophy of minimal setup friction.