swictation
v0.7.35
Published
Cross-platform voice-to-text dictation for Linux and macOS with GPU acceleration (NVIDIA CUDA/CoreML), Secretary Mode (60+ natural language commands), Context-Aware Meta-Learning, and pure Rust performance. Meta-package that automatically installs platfor
Maintainers
Readme
Swictation
Voice-to-text dictation for Linux and macOS with GPU acceleration. Pure Rust daemon with VAD-triggered auto-transcription, sub-second latency, and complete privacy.
Supported platforms:
- Linux x64 (Ubuntu 24.04+, GLIBC 2.39+)
- macOS Apple Silicon (M1/M2/M3/M4/M5), macOS 14 Sonoma+
Install
npm install -g swictation --foreground-scriptsThe --foreground-scripts flag shows installation progress. Postinstall automatically:
- Detects your GPU and downloads optimized acceleration libraries
- Downloads and test-loads AI models (~30-60s on first install)
- Sets up the background service (systemd on Linux, launchd on macOS)
- Shows platform-specific setup instructions
Platform Requirements
Linux x64
- Ubuntu 24.04+ (GLIBC 2.39+), Node.js 18+
- Optional: NVIDIA GPU with 4GB+ VRAM for CUDA acceleration (CPU fallback available)
macOS Apple Silicon (M1-M5)
- macOS 14 Sonoma or later, Node.js 18+
- 8GB+ RAM (16GB+ recommended for the 1.1B model)
- Accessibility permissions granted during setup
Commands
swictation start # Start the daemon
swictation stop # Stop the daemon
swictation status # Show service status
swictation toggle # Toggle recording on/off
swictation --version # Show versionDocumentation
Full documentation, configuration reference, and troubleshooting guides: https://github.com/robertelee78/swictation
Links
- Source: https://github.com/robertelee78/swictation
- Issues: https://github.com/robertelee78/swictation/issues
License
Apache-2.0
