Labelingo

Annotate UI screenshots with translations.

Features

• 🔍 Detects text using OCR services (OpenCV, Tesseract)
• 🌐 Translates using AI services (OpenAI, Anthropic Claude)
• 🎨 Generates interactive SVG with annotations and translations
• 📁 Supports batch processing of multiple files

Quick Start

1. Install:

pip install labelingo

1. Set up your API keys:

export OPENAI_API_KEY=your-openai-api-key
export ANTHROPIC_API_KEY=your-anthropic-api-key  # Optional, for Claude

1. Process an image:

labelingo screenshot.png            # Basic usage
labelingo screenshot.png -l fr      # Translate to French
labelingo *.png -o translated/     # Process multiple files

Installation Options

For additional OCR service support:

pip install 'labelingo[ocr]'       # Install OCR service dependencies

For PNG/PDF output:

pip install 'labelingo[cairo]'     # Install Cairo dependencies

For all features:

pip install 'labelingo[ocr,cairo]' # Install all optional dependencies

See System Requirements for platform-specific dependencies.

Documentation

• Command Line Reference - Detailed usage and examples
• Design Document - Architecture and design decisions
• Development Guide - Setup and contribution guidelines

License

MIT License. See LICENSE file.