🤖 Jarvis – Personal AI Assistant

Jarvis is a Python-based personal AI assistant that supports voice and text interaction, system automation, and intelligent command handling.

It is designed as a learning-focused, extensible desktop assistant, inspired by real-world AI systems.

🚧 Status: Actively under development

---

✨ Features

• 🗣️ Voice-based interaction (Speech Recognition + Text-to-Speech)
• 💬 Text input via GUI
• 🎯 Wake-word activation (Jarvis)
• 🔁 Continuous conversation mode (no repeated wake word)
• 🧠 Intent detection with LLM fallback
• ⚙️ System & browser automation
• 🧱 Modular and clean architecture
• 🔐 Secure API key handling using environment variables

---

🛠️ Tech Stack

Language: Python 3

Libraries & Tools:

• SpeechRecognition
• pyttsx3
• Tkinter (GUI)
• OpenAI / Gemini (optional LLM integration)

Version Control: Git & GitHub

---

📂 Project Structure (Simplified)

jarvis/
├── main.py
├── config.py
├── voice.py
├── intent.py
├── automation.py
├── llm.py
├── gui.py
├── requirements.txt
├── .gitignore
└── data/

---

🔐 Environment Setup

API keys are never hardcoded.

Create a .env file in the project root:

OPENAI_API_KEY=your_openai_api_key_here
GEMINI_KEY_PATH=path_to_gemini_json_key

Ensure .env is added to .gitignore to keep secrets safe.

---

▶️ How to Run

pip install -r requirements.txt
python main.py

Requirements

• 🎤 Microphone (for voice mode)
• 🌐 Internet connection (for LLM-based features)

---

⚠️ Known Limitations

• Memory system is basic (no long-term memory yet)
• Intent detection is currently rule-based
• Limited built-in skills
• Windows-focused automation

---

🗺️ Roadmap

• Persistent memory (JSON / SQLite)
• Plugin-based skill system
• Improved intent classification
• Personality modes
• Offline command support
• Packaging as executable (.exe)

---

🤝 Contributing

This is a personal learning project.

Suggestions, issues, and pull requests are welcome and appreciated.

---

👤 Author

Mohit Rajput GitHub: https://github.com/msrajput08