Open Source AI Powered voice to text dictation tool, alternative to superwhisper, whispr flow
Download • Features • Installation • Usage
VoiceTypr is an open source AI voice-to-text dictation tool, alternative to Wispr Flow and SuperWhisper. Available for macOS and Windows. Pay once, use forever.
- System-wide hotkey for quick recording
- Automatic text insertion at cursor position
- Works in any app - cursor, claude code, chatgpt, slack, etc
- 100% offline transcription - your voice never leaves your device
- Multiple model sizes for accuracy vs speed tradeoffs
- Support for 99+ languages out of the box
- Hardware acceleration (Metal on macOS)
- Built with Rust and Tauri for blazing-fast performance
- Optimized for each platform with hardware acceleration
- Minimal resource usage with maximum efficiency
- Complete offline operation - no cloud, no tracking (only trial check)
- Your recordings stay on your device
- Open source for full transparency
- Transform your transcriptions with AI (Groq/Gemini)
- Smart presets: Prompts, Email, Commits, Notes
- Secure API key storage
- Requires internet connection for enhancement only
- Clean, user interface
- Menubar integration for quick access
- Visual feedback during recording
- Auto-updates to keep you on the latest version
- macOS 13.0 (Ventura) or later
- 3-4 GB free disk space (for AI models)
- Microphone access permission
- Accessibility access permission
- Windows 10/11 (64-bit)
- 3-4 GB free disk space (for AI models)
- GPU acceleration available (5-10x faster with NVIDIA, AMD, Intel GPUs)
- Download the latest VoiceTypr.dmg
- Open the DMG and drag VoiceTypr to Applications
- Launch VoiceTypr from Applications
- Follow the onboarding to download your preferred AI model
Note: VoiceTypr is fully signed and notarized by Apple, so you can run it without security warnings.
- Download the latest VoiceTypr installer
- Run the installer
- Launch VoiceTypr from Start Menu
- Follow the onboarding to download your preferred AI model
GPU Acceleration (5-10x faster)
- VoiceTypr automatically uses your GPU if available
- For best performance, ensure your graphics drivers are up to date:
- Falls back to CPU automatically if GPU unavailable
- Launch VoiceTypr - Find it in your Applications folder (macOS) or Start Menu (Windows)
- Grant Permissions - Allow microphone access (and accessibility on macOS)
- Download a Model - Choose from tiny to large models based on your needs
- Start Transcribing - Press your hotkey anywhere to record
- 🎯 Quick Cancel: Double Press
Escwhile recording to cancel - 📝 Long Recordings: VoiceTypr handles extended recordings seamlessly but shorter recordings are recommended to do.
- 🌍 Multiple Languages: Just speak - Whisper auto-detects the language
- ⚡ Instant Insert: Text appears right where your cursor is
voicetypr/
├── src/ # React frontend
│ ├── components/ # UI components
│ ├── hooks/ # Custom React hooks
│ └── types/ # TypeScript types
├── src-tauri/ # Rust backend
│ ├── src/
│ │ ├── audio/ # Audio recording
│ │ ├── whisper/ # Whisper integration
│ │ └── commands/ # Tauri commands
│ └── capabilities/ # Security capabilities
├── scripts/ # Build and utility scripts
└── tests/ # Test suites
VoiceTypr automatically detects and uses your GPU for faster transcription. If you're experiencing slower performance:
Update your graphics drivers - This is the most common fix:
Note: VoiceTypr always works - it automatically falls back to CPU if GPU acceleration is unavailable
VoiceTypr is licensed under the GNU Affero General Public License v3.0.