Manhwa Translation Tool

🛠️ Run Project

• Frontend: cd frontend && python -m http.server 3000
• Backend: cd backend &&

A comprehensive admin interface for streamlining manhwa translation workflows, built with Supabase, FastAPI, and Manga OCR.

🚀 Features

• Series Management: Add, edit, and manage manhwa series
• Chapter Processing: Upload chapters with automatic OCR text extraction
• AI-Powered Translation: GPT-4/DeepL integration with translation memory
• Quality Assurance: AI insights and feedback loops
• Visual Glossary: Character and term cards with summaries
• Role-Based Access: Admin, editor, and translator roles

🏗️ Architecture

• Backend: Supabase (PostgreSQL + Auth + Storage)
• OCR Service: Manga OCR via FastAPI
• Translation: OpenAI GPT-4 / DeepL API
• Admin Interface: Supabase Admin UI
• Storage: Supabase Buckets for images

📋 Prerequisites

• Python 3.8+
• Node.js 16+
• Supabase account
• OpenAI API key (optional)
• DeepL API key (optional)

🛠️ Setup

1. Clone and Install Dependencies

git clone <repository-url>
cd manhwatrans

2. Backend Setup (FastAPI)

cd backend
pip install -r requirements.txt

3. Environment Configuration

Create .env files in both backend/ and frontend/ directories:

# Backend .env
SUPABASE_URL=your_supabase_url
SUPABASE_ANON_KEY=your_supabase_anon_key
SUPABASE_SERVICE_ROLE_KEY=your_service_role_key
OPENAI_API_KEY=your_openai_key
DEEPL_API_KEY=your_deepl_key

4. Database Setup

Run the SQL scripts in database/ to set up your Supabase tables.

5. Start Services

# Start FastAPI backend
cd backend
uvicorn main:app --reload

# Start frontend (if using custom UI)
cd frontend
npm run dev

📊 Database Schema

The system uses the following core tables:

• series - Manhwa titles
• chapters - Chapter containers
• pages - Individual images
• text_boxes - OCR detected text regions
• translation_memory - Approved translations
• dialogue_history - Previous dialogues
• ai_glossary - Character/term cards
• feedback_log - Editor feedback
• users - User management

🔄 Workflow

1. Upload Chapter: Add images to a chapter
2. OCR Processing: Automatic text extraction via Manga OCR
3. AI Translation: Generate translations with context
4. Editor Review: Manual corrections and feedback
5. Quality Check: AI insights and suggestions
6. Publish: Final review and publication

🤝 Contributing

Please read CONTRIBUTING.md for details on our code of conduct and the process for submitting pull requests.

📄 License

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