Augen

AI-Powered Vision Assistant for Accessibility

A comprehensive accessibility-focused vision assistant featuring voice interaction, multilingual support, and intelligent image analysis. Built with modern web technologies and designed for users with visual impairments.

Features

Core Functionality

• Three-Mode UX System: Chat Mode, Image Mode, and Voice+Image Mode
• AI-Powered Vision: Advanced image analysis using Groq Llama Scout 4
• Voice Interaction: Real-time transcription and bidirectional voice chat
• Text-to-Speech: High-quality TTS with stop controls and number formatting
• Multilingual Support: 12 languages with automatic browser detection

Multilingual Support

• English, Spanish, French, German, Italian, Portuguese
• Russian, Japanese, Korean, Chinese, Arabic, Hindi
• Auto-detection from browser language preferences
• Complete UI internationalization (i18n)

Accessibility Features

• Screen reader optimized interface
• High contrast mode support
• Material Design accessibility standards
• Voice-first interaction design
• Keyboard navigation support

Technical Features

• Backend: Cloudflare Workers with Groq API integration
• Frontend: Vanilla JavaScript with Material Design
• Voice Processing: Groq Whisper for transcription
• Security: Input sanitization and CORS protection
• Performance: Intelligent image compression and caching

Quick Start

Prerequisites

• Node.js (v18 or later)
• Cloudflare account
• Groq API key

Installation

1. Clone the repository

   git clone https://github.com/sortedmess/augen.git
   cd augen

2. Install dependencies

   npm install

3. Setup environment

   npm run setup
   # This copies wrangler.toml.local to wrangler.toml

4. Configure your Groq API key

   npx wrangler secret put GROQ_API_KEY -e production

5. Deploy the worker

   npm run deploy:worker

6. Start local development

   npm run serve

Visit http://localhost:8080 to see the application running locally.

Architecture

Frontend

• Vanilla JavaScript → No frameworks, maximum compatibility
• Static hosting → Deploy anywhere (GitHub Pages, Netlify, Vercel)
• Mobile-first → Optimized for touch devices
• High contrast CSS → WCAG 2.1 AA compliant

Backend

• Cloudflare Workers → Serverless, global edge deployment
• Groq API integration → Deploy your own worker with your Groq API key
• Groq APIs → Ultra-fast AI inference
  • Vision: Llama Scout 4
  • Voice: Whisper for transcription
  • TTS: Native browser speech synthesis (offline, no API cost)

API Endpoints (Deploy Your Own)

After deploying your own Cloudflare Worker, you'll have:

• POST /api/analyze - Vision analysis endpoint
• POST /api/transcribe - Voice transcription endpoint
• POST /api/voice-query - Voice chat endpoint
• GET /api/health - Health check endpoint

Note: You need to deploy your own worker with your Groq API key for production use

Cost-Effective & Accessible

• No TTS API costs → Uses free native browser speech synthesis
• ~90% cheaper vision API than OpenAI GPT-5 with competitive performance that beats GPT-4o
• Live demo available → See the full app in action at the demo site
• Groq API required → Deploy your own Cloudflare Worker with your Groq API key
• Open source → Full code available for customization and deployment

Use Cases

Documents & Text

• Restaurant menus
• Street signs
• Product labels
• Documents and papers
• Handwritten notes

Scene Description

• Navigation assistance
• Object identification
• People and activities
• Environmental awareness
• Safety assessment

Customizable

• Morse code output for deaf-blind users
• Haptic feedback patterns
• Summary vs detailed descriptions
• Multiple language support

Development

Project Structure

augen/
├── assets/
│   ├── css/style.css          # Material Design styling
│   └── js/script.js           # Main application logic
├── src/
│   └── worker/
│       └── worker.js          # Cloudflare Workers backend
├── index.html                 # Main application interface
├── wrangler.toml             # Cloudflare Workers config
└── package.json              # Project metadata

Local Development

# Start local server
npm run serve

# Start worker development
npm run dev

# Deploy to production
npm run deploy

Contributing

We welcome contributions! This project is built for the accessibility community.

Areas for Contribution

• Accessibility improvements → Better screen reader support, motor accessibility
• Language support → Multi-language interface and descriptions
• Voice options → Different TTS voices and speeds
• Performance → Faster processing, offline capabilities
• Documentation → Better guides, tutorials, examples

Getting Started

1. Fork the repository
2. Create a feature branch: git checkout -b feature/amazing-feature
3. Make your changes
4. Test accessibility with screen readers
5. Submit a pull request

License

This project is licensed under the AGPL-3.0 License - see the LICENSE file for details.

Support

• Documentation → Help page
• Issues → GitHub Issues
• Discussions → GitHub Discussions

---

Built for accessibility

Augen means "eyes" in German - helping everyone see the world through AI.