🚀 OpenFlow

A macOS-native, GPU-accelerated, streaming speech-to-text menu bar app.

---

OpenFlow is an open-source, local-first alternative to Wispr Flow. It provides instant voice dictation into any application using Apple Silicon's Neural Engine (via MLX) for ultra-low latency transcription.

✨ Features

• ⚡️ Ultra-Low Latency: Optimized for Apple Silicon using MLX (Metal) and faster-whisper.
• 🔒 Private & Local: All processing happens on-device. No audio ever leaves your Mac.
• 🎙️ Voice-Reactive UI: Beautiful, floating overlay that responds to your voice in real-time.
• ⌨️ Universal Dictation: Inserts text directly into any active application via Accessibility APIs.
• 🔇 Smart Denoising: Integrated RNNoise for crystal-clear audio capture even in noisy environments.
• 📦 Easy Installation: Simple manual setup.

📥 Installation

Manual Build

Requires Python 3.11+ and portaudio.

1. Clone the repository:

   git clone https://github.com/sammy4321/OpenFlow.git
   cd OpenFlow

2. Run the installer:

   # This script sets up a virtual environment, compiles dependencies, and downloads models.
   make install

3. Run the app:

   make run

⚙️ Configuration

Changing Models

By default, OpenFlow uses the tiny model for maximum speed (<300ms). You can switch to larger models for better accuracy:

# Run with a specific model size
make run MODEL=base
make run MODEL=small
make run MODEL=medium

Supported models: tiny, base, small, medium, large-v3. Functionality depends on your machine's memory.

🤝 Contributing

We welcome contributions! Please follow these steps:

1. Fork the repository.
2. Create feature branch (git checkout -b feature/AmazingFeature).
3. Commit your changes (git commit -m 'Add some AmazingFeature').
4. Push to the branch (git push origin feature/AmazingFeature).
5. Open a Pull Request.

📄 License

Distributed under the MIT License. See LICENSE for more information.