Transform your document workflow without compromising security. SnapDoc AI processes everything on-device, ensuring your sensitive information never leaves your control.
- Your Data Stays Yours: In a world where data security is paramount, SnapDoc AI never sends your documents to external servers
 - Work Uninterrupted: Process documents even without internet connection - perfect for secure environments and travel
 - Enterprise Confidence: Meet compliance requirements with auditable, on-device processing
 - Natural Interaction: Control the application with voice commands and listen to documents - all processed locally
 - Accessibility Built-in: Voice commands and text-to-speech make documents accessible to everyone in your organization
 
- Intelligent Document Analysis: Instantly extract key information and insights
 - Powerful Summarization: Save hours by generating concise summaries of lengthy documents
 - Smart Information Extraction: Automatically identify dates, names, amounts, and contractual terms
 - Voice Command Control: Navigate and analyze documents hands-free with natural voice instructions
 - Text-to-Speech Conversion: Listen to your documents while multitasking
 - Multi-Format Support: Seamlessly handle PDF, Word, plain text, and images (with on-device OCR)
 - Enterprise-Ready Architecture: Built with structured logging, flexible configuration, and comprehensive error handling
 
SnapDoc AI leverages cutting-edge AI technology that runs entirely on your device:
- Qualcomm AI Hub models optimized for Snapdragon devices
 - On-device machine learning for document understanding
 - Privacy-preserving voice recognition
 - ONNX Runtime with DirectML acceleration
 - Modern software architecture following enterprise best practices:
- Structured logging with contextual information
 - Flexible configuration management
 - Comprehensive error handling
 - Extensive unit testing coverage
 
 
This application demonstrates the power of on-device AI using:
- Qualcomm AI Hub models
 - Snapdragon X series processors
 - Windows on Arm optimization
 - ONNX Runtime with DirectML acceleration
 - Modern Python best practices:
- Structured logging with structlog
 - Configuration management with dynaconf
 - Comprehensive error handling
 - Unit testing with pytest
 
 
# Clone the repository
git clone https://github.com/devraftel/snapdoc-edge-ai
cd snapdoc-edge-ai
# Install with uv
uv pip install -e .
# For development (includes testing tools)
uv pip install -e ".[dev]"
# Run the application
python -m snapdoc_ai# Create and activate a virtual environment
python -m venv venv
source venv/bin/activate  # On Unix/Mac
venv\Scripts\activate  # On Windows
# Install dependencies
pip install -e .
# Run the application
python -m snapdoc_aiThe project includes a Makefile with common development commands:
# Install development dependencies
make dev-install
# Run the application
make run
# Run tests
make test
# Run tests with coverage
make coverage
# Format code
make format
# Lint code
make lint
# Clean build artifacts
make cleanSnapDoc AI can be configured in several ways:
- Default settings: Located in 
src/snapdoc_ai/default_settings.toml - User settings: Create a 
settings.tomlfile in~/.snapdoc-ai/ - Environment variables: Set variables with 
SNAPDOC_prefix (e.g.,SNAPDOC_DEBUG=true) 
- Windows on Snapdragon device (or any modern system for development)
 - Python 3.11+
 - For OCR: Tesseract OCR installed (for image-based documents)
 
MIT License