README.md

metadata

title: Multi-Lingual Product Catalog Translator
emoji: 🌐
colorFrom: blue
colorTo: green
sdk: streamlit
sdk_version: 1.28.0
app_file: app.py
pinned: false
license: mit
tags:
  - translation
  - indictrans2
  - multilingual
  - ai4bharat
  - indian-languages
  - neural-machine-translation
  - ecommerce
  - product-catalog
short_description: AI-powered translator for Indian languages using IndicTrans2

Multi-Lingual Product Catalog Translator 🌐

AI-powered translation service for e-commerce product catalogs using IndicTrans2 by AI4Bharat.

🚀 Quick Start - One Command Deployment

Universal Deployment (Works on Any Platform)

# Clone and deploy in one command
git clone https://github.com/your-username/multilingual-catalog-translator.git
cd multilingual-catalog-translator
chmod +x deploy.sh
./deploy.sh

Platform-Specific Deployment

macOS/Linux

./deploy.sh                    # Auto-detect best method
./deploy.sh docker             # Use Docker
./deploy.sh standalone         # Without Docker

Windows

deploy.bat                     # Auto-detect best method
deploy.bat docker              # Use Docker
deploy.bat standalone          # Without Docker

Cloud Platforms

./deploy.sh hf-spaces          # Hugging Face Spaces
./deploy.sh cloud railway      # Railway
./deploy.sh cloud render       # Render
./deploy.sh cloud heroku       # Heroku

---

Multi-Lingual Product Catalog Translator

Real AI-powered translation system for e-commerce product catalogs supporting 15+ Indian languages with neural machine translation powered by IndicTrans2 by AI4Bharat.

🚀 Features

• 🤖 Real IndicTrans2 AI Models - 1B parameter neural machine translation
• 🌍 15+ Languages - Hindi, Bengali, Tamil, Telugu, Malayalam, Gujarati, and more
• 📝 Product Catalog Focus - Optimized for e-commerce descriptions
• ⚡ GPU Acceleration - Fast translation with Hugging Face Spaces GPU
• 🎯 High Accuracy - State-of-the-art translation quality

🌍 Supported Languages

English, Hindi, Bengali, Gujarati, Kannada, Malayalam, Marathi, Odia, Punjabi, Tamil, Telugu, Urdu, Assamese, Nepali, Sanskrit

🏗️ Technology

• AI Models: IndicTrans2-1B by AI4Bharat
• Framework: Streamlit + PyTorch + Transformers
• Deployment: Hugging Face Spaces with GPU support
• Languages: Real neural machine translation (not simulated)

🎯 Use Cases

• E-commerce product localization for Indian markets
• Multi-language content creation
• Educational and research applications
• Cross-language communication tools

🙏 Acknowledgments

• AI4Bharat for the amazing IndicTrans2 models
• Hugging Face for providing free GPU hosting
• Streamlit for the web framework