mojo-voice
COMPILING _Lightning-fast, privacy-first voice dictation for developers
Overview
mojo-voice is a lightning-fast, privacy-first voice dictation tool built specifically for developers. Press a hotkey, speak naturally, and your words appear instantly at your cursor position—no cloud, no latency, no compromises.
Perfect for writing documentation, drafting emails, or dictating code comments without breaking flow.
⚠️ Active Development
This project is under active development. It's usable locally but not yet polished for general distribution. Expect frequent updates and breaking changes.
Key Features
- Fully local processing - no cloud, no data leaves your machine
- Sub-100ms latency from speech to text
- Global hotkey activation
- Automatic cursor injection
- Works in any application
- Open-source and free
Installation
Currently in early development. You can clone and run locally if you're comfortable with:
- Building from source
- Configuring system hotkeys
- Troubleshooting audio input issues
Check the GitHub repository for setup instructions.
Tech Stack
- Mojo for performance-critical transcription engine
- Whisper.cpp for local speech recognition
- System-level hotkey integration
- Zero network dependencies
Roadmap
- Homebrew/Go installation (coming soon)
- Configuration GUI
- Multi-language support
- Custom command macros
- Plugin system for IDE integration
Performance
Built with Mojo for C-level performance with Python's ease of use. On modern hardware, transcription happens faster than you can speak, with negligible system overhead.
No cloud latency, no API rate limits, no privacy concerns. Everything runs locally.