A highly advanced AI assistant with a real-time particle globe UI, speech interaction, and system integration.
- Real-time Particle UI: Dynamic particle sphere that reacts to your voice.
- Voice Interaction: Wakes up with "Hey Jarvis" or "Jarvis".
- Speech-to-Text: High-accuracy STT using Faster Whisper (
tiny.en). - Text-to-Speech: Smooth, human-like responses with Microsoft Edge TTS.
- AI Brain: Integrated with OpenRouter (online) or Ollama (local models).
- System Integration: Opens applications, searches the web, and creates files.
- Real-time Monitoring: Tracks CPU, RAM, and Battery via internal sensors.
- Python 3.12 or higher.
- FFmpeg (required for audio processing).
- A microphone and speaker.
git clone https://github.com/Linuxthecoder/JARVIS.git
cd JARVISUsing uv (recommended):
uv pip install -r requirements.txtOr with standard pip:
pip install -r requirements.txtCreate a .env file in the root directory and add your keys:
LLM_PROVIDER=api
OPENROUTER_API_KEY=your_key_here
OPENROUTER_MODEL=meta-llama/llama-3.1-8b-instruct:free
# Optional: Ollama settings
# LLM_PROVIDER=local
# OLLAMA_BASE_URL=http://localhost:11434
# OLLAMA_MODEL=llama3Start the integrated UI and brain with a single command:
uv run jarvis_ui.pyOr:
python jarvis_ui.py- Manual Wake: Click the "WAKE JARVIS" button in the control panel.
- Sphere Interaction: Click the particle globe to open the file explorer.
- Language Switch: Use the UI to switch between English, Hindi, and Telugu.
- Chat History: Click the "CHAT HISTORY" button to view recent interactions.
jarvis_ui.py: The main entry point and UI controller (Integrated HTML/JS/CSS).jarvis_brain.py: Core AI logic, STT/TTS processing, and system commands.requirements.txt: Project dependencies list..env: Environment configuration and API keys (ignore in Git).
- Interrupt Jarvis: Say "Stop", "Shut up", or "Enough" to silence him immediately.
- Go to Sleep: Say "Go to sleep" to hibernate the active listener.