🚀 A Futuristic AI-Powered Personal Assistant Inspired by J.A.R.V.I.S.

VYOM is built on a Modular Multi-Threaded Architecture. Unlike linear assistants, VYOM decouples peripheral I/O (Voice/Listen) from core logic (NLP/Action) to prevent UI freezing and ensure real-time responsiveness.

The following diagram illustrates how a voice command propagates through the modular layers:

graph TD
    subgraph Input_Layer [Perception]
        A[🎤 Voice Input] -->|PyAudio / SpeechRecognition| B(Audio Stream)
        B -->|Whisper / Google API| C{Speech-to-Text}
    end

    subgraph Brain_Layer [Processing]
        C -->|Raw Text| D[🧠 NLP Engine]
        D -->|Intent Extraction| E{Action Router}
    end

    subgraph Execution_Layer [Action]
        E -->|System Cmd| F[OS Controller]
        E -->|Web Query| G[Browser Automation]
        E -->|API Call| H[Weather/IoT/News]
    end

    subgraph Output_Layer [Feedback]
        F & G & H --> I[🗣️ TTS Engine]
        I --> J[🔊 Speaker Output]
    end

To maintain the "Always Listening" capability while executing heavy AI tasks, VYOM utilizes Python's threading and asyncio modules:

• Thread 1 (Listener): Continuously monitors the microphone for the wake word.
• Thread 2 (Processor): Handles API calls to Groq/Cohere without blocking the listener.
• Thread 3 (Executor): Manages OS-level tasks and GUI updates.

For SWOC contributors, please refer to this modular map before submitting PRs:

VYOM/
├── core/
│   ├── engine.py      # Main loop & multi-threading orchestration
│   ├── listener.py    # Voice capture & STT logic
│   └── speaker.py     # TTS implementation
├── modules/           # Modular skills (Add new features here)
│   ├── system_ops.py  # File handling & OS controls
│   └── web_search.py  # Playwright/Selenium automation
├── docs/              # Detailed technical documentation
├── data/              # Model weights, logs, and user configurations
└── main.py            # Entry point for the application

• Python 3.13+
• FFmpeg (Required for audio processing)
• C++ Build Tools (Required for PyAudio on Windows)

🐧 Linux/Mac Setup (Audio Dependencies) Most setup errors occur due to missing audio driver headers. Run the following before pip install:

• For Ubuntu/Debian:

sudo apt-get update
sudo apt-get install python3-pyaudio portaudio19-dev libasound2-dev espeak

• For macOS:

brew install portaudio
pip install pyaudio

1. Clone & Environment

git clone [https://github.com/th-shivam/vyom.git](https://github.com/th-shivam/vyom.git) && cd vyom
python -m venv .venv
source .venv/bin/activate  # Mac/Linux
# .venv\Scripts\activate   # Windows

2. Install & Run

pip install -r requirements.txt
python main.py

We are proud to be an official part of Social Winter of Code (SWOC) 2026! 🚀

We welcome contributors of all skill levels. To ensure a smooth collaboration, please identify your path:

• 🌱 Beginners: Look for issues labeled good-first-issue and documentation. Perfect for your first PR!
• 🛠️ Advanced: Check for modular-enhancement and threading-optimization to work on the core engine.

1. Fork the repository and create your branch.
2. Follow the PEP 8 style guide for Python code.
3. Ensure your module is placed in the correct directory (see Project Structure).
4. Open a PR with a clear description of your changes.

📋 Full Contributing Guide | 🏗️ Architecture Deep Dive

This project is licensed under the MIT License. You are free to use, modify, and distribute this software, provided the original copyright and license notice are included.

TL;DR: Open-source, permissive, and community-friendly.

See the LICENSE file for the full legal text.