Beacon is a lightweight AI-powered desktop assistant that listens to your voice, executes commands, and assists you with daily tasks.
It uses Whisper.cpp for speech recognition and integrates automation features to make your workflow faster and more accessible.
- 🎙️ Wake Word Activation → Say "Hey Beacon" to start listening
- 🗣️ Voice Commands → Perform tasks hands-free
- 📸 Screenshot Capture → Automatically take desktop screenshots
- 📂 Cross-Platform Support → Works on Windows, Linux, and macOS
- ⚡ Optimized Performance → Runs efficiently on your CPU with full resource utilization
- 🛠️ Customizable Shortcuts → Desktop shortcut & Start Menu integration
- 📜 License Window → Displays license agreement during installation
- Download the Beacon Installer (SFX Archive)
- Run the installer and follow the setup instructions
- The installer will:
- Extract required files
- Run
windows.bat
automatically - Create a desktop shortcut for
Beacon.exe
- Say "Hey Beacon" → Assistant activates
- Speak your command clearly → Example: "Take screenshot"
- Beacon executes the action and gives feedback
- Screenshots are saved in the default folder (
G:\Assistant\screen.png
by default)
Beacon/
│── beacon/ # Core assistant files
│── models/ # AI/Whisper models
│── whisper.cpp/ # Speech-to-text engine
│── Beacon.exe # Main application
│── accuracy.py # Accuracy testing script
│── speed.py # Performance testing script
│── windows.bat # Windows startup script
│── Mac.bat # MacOS startup script
│── Linux.bat # Linux startup script
│── requirements.txt # Dependencies
│── LICENSE.txt # License agreement
│── gemini_key.txt # API key storage
│── logo.ico # Application icon
│── screen.png # Screenshot output
- OS: Windows / Linux / MacOS
- CPU: Intel i5-2400 or higher (Beacon is optimized to use full CPU power)
- RAM: 8 GB minimum (16 GB recommended)
- Dependencies: Listed in
requirements.txt
Install dependencies with:
pip install -r requirements.txt
Beacon is designed with scalability in mind. In future updates, the assistant will be enhanced with:
- Smarter Voice Recognition using advanced ML models.
- Multi-language Support for a wider audience.
- Plugin System to allow developers to integrate custom commands.
- Cross-Platform Compatibility (Windows, Linux, macOS).
- Accessibility Features for people with disabilities, ensuring inclusivity.
Follow these steps to set up and run Beacon on your system:
- Clone the Repository
git clone https://github.com/YourUsername/beacon.git cd beacon
- Set Up Virtual Environment (Recommended)
python -m venv venv source venv/bin/activate # On Linux/Mac venv\Scripts\activate # On Windows
- Install Dependencies
pip install -r requirements.txt
- Run Beacon
python beacon.py
- Beacon will now listen for your commands and respond through voice + automation features.
- The accuracy.py file helps improve the command recognition accuracy by:
- Listening to Input – Captures your microphone input via Whisper.cpp.
- Processing the Text – Converts speech into structured text commands.
- Matching Commands – Compares spoken commands with the pre-defined dictionary.
- Example: “open browser” → maps to open_browser() function.
- Error Handling – If input is unclear, it prompts you to repeat.
- Logging Results – Stores recognition results for accuracy analysis and debugging.
- This ensures Beacon doesn’t just “hear” you but also understands you correctly.
- We are already planning Beacon V2.0, with a focus on accessibility and real-world utility:
- Blind-Friendly Mode:
- Full voice feedback for every action.
- Auto-reading system messages, errors, and outputs.
- Hands-free navigation support.
- Smart Task Scheduling: Set reminders, timers, and voice-controlled notes.
- IoT Integration: Control smart devices like lights, fans, or appliances.
- Improved Context Awareness: Beacon will remember past commands for a smoother conversation.
- 📧 Issac Moses D – issacmoses19082005@gmail.com
- 📧 Varsha R – Varsharajmohan236@gmail.com
- 💼 LinkedIn 👉 Stay tuned for release updates in the Releases tab of this repo!