LegalDoc-Translate-Query-Assistant

Overview

LegalDoc-Translate-Query-Assistant is a powerful AI-driven legal document processing tool designed to:

• Translate legal case PDFs in any language to English.
• Store translated content as embeddings in Weaviate.
• Provide a RAG-based LLM assistant to answer legal queries using Groq's LLM.

The application consists of:

• Backend: A FastAPI server that handles translation, embedding storage, and query processing.
• Frontend: A React.js web app built with TypeScript and Material UI for seamless interaction.

---

Features

PDF Translation - Convert any legal case document into English. Vector Search with Weaviate - Efficient retrieval of relevant legal information. RAG-based LLM Assistant - Provides evidence-based legal answers. FastAPI Backend - Scalable and efficient API services. Frontend UI - Built with React.js and Material UI for a smooth user experience.

---

Technologies Used

Component	Technologies Used
Backend	FastAPI, PyTorch, MBart, Weaviate, SentenceTransformers, Groq LLM
Frontend	React.js, TypeScript, Material UI
Database	Weaviate Vector DB

---

Project Structure

LegalDoc-Translate-Query-Assistant
│── backend
│   ├── app.py                  # FastAPI server
│   ├── rag_pipeline.py          # Retrieval-Augmented-Generation pipeline
│
│── frontend
│   ├── src
│   │   ├── components           # UI Components
│   │   ├── pages
│   │   │   ├── Home.tsx         # PDF Translation Page
│   │   │   ├── Assistant.tsx    # Query Assistant Page
│   │   ├── App.tsx              # Main React Component
│   │   ├── main.tsx             # React entry point
│
│── .env                         # Environment variables
│── README.md                    # Documentation

---

Environment Setup

Prerequisites

Ensure you have:

• Anaconda installed.
• A compatible NVIDIA GPU (for CUDA acceleration).
• Python 3.9.

Step 1: Create and Activate Conda Environment

conda create -n legal-rag python=3.9 -y
conda activate legal-rag

Step 2: Install PyTorch with CUDA Support

pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu121

Step 3: Install Required Libraries

pip install transformers fastapi uvicorn python-magic PyPDF2 sentence-transformers weaviate-client python-dotenv groq

Step 4: Install Additional Dependencies

pip install tiktoken protobuf python-multipart

Step 5: Upgrade Key Libraries

pip install --upgrade transformers sentencepiece

Verification

Run the following to confirm successful installation:

python -c "import torch; print(torch.__version__)"
python -c "import transformers; print(transformers.__version__)"
python -c "import torch; print(torch.cuda.is_available())"

---

Running the Application

Backend (FastAPI)

cd backend
python app.py

Frontend (React.js)

cd frontend
npm install
npm run dev

The application will be available at http://localhost:5173/.

---

Usage Guide

1. Upload a Legal Case PDF (Home Page)

• Upload a document in any language.
• The backend will translate it into English.
• Translated content is stored in Weaviate for future retrieval.

2. Query the Legal Assistant (Assistant Page)

• Ask legal questions related to the case.
• The assistant retrieves relevant context from Weaviate.
• It generates answers ONLY based on the provided case PDF.

---

Results

Home Page - PDF Translation

Assistant Page - Query Assistant

---