R1 Supervisor Multimodal AI Assistant

A multimodal AI assistant powered by the R1 Supervisor system, capable of processing text, audio, and visual inputs.

Features

• 🎤 Audio Processing

  • Speech recognition
  • Text-to-speech synthesis
  • Voice command processing

• 🖼️ Visual Processing

  • Image generation
  • Image analysis
  • Visual content understanding

• 🔍 Information Retrieval

  • Web search integration
  • Wikipedia queries
  • Link scraping and processing

• 📄 File Handling

  • Multi-format file processing
  • Document analysis
  • Data extraction

Installation

# Clone the repository
git clone https://github.com/Scruff-AI/r1s-multimodal.git
cd r1s-multimodal

# Create virtual environment
python -m venv venv

# Activate virtual environment
# On Windows:
.\venv\Scripts\activate
# On Unix or MacOS:
source venv/bin/activate

# Install dependencies
pip install -r requirements.txt

Project Structure

r1s-multimodal/
├── src/
│   ├── core/          # Core system functionality
│   ├── audio/         # Audio processing modules
│   ├── vision/        # Visual processing modules
│   └── utils/         # Utility functions
├── tests/             # Test suite
├── docs/              # Documentation
├── .gitignore         # Git ignore rules
├── README.md          # Project documentation
└── requirements.txt   # Project dependencies

Environment Setup

Create a .env file in the root directory with the following variables:

# API Keys and Configuration
OPENAI_API_KEY=your_key_here
CHAINLIT_AUTH_SECRET=your_secret_here

# Model Settings
MODEL_NAME=your_model_name
DEVICE=cuda  # or cpu

License

MIT License