Omani Legal AI Assistant

A bilingual Legal AI Assistant specialized in Omani legal documentation, offering advanced document processing capabilities across Arabic and English languages.

Features

• Document Q&A: Ask questions about legal documents in English or Arabic
• Article Summarization: Get concise summaries of specific legal articles
• Law Comparison: Compare different laws to understand similarities and differences
• Case Analysis: Submit a case description and get legal analysis based on relevant laws
• Document Creation: Generate legal documents based on user specifications
• Voice Input: Upload voice recordings for hands-free interaction
• AI-Powered Text Improvement: Enhanced readability for legal citations
• PDF Generation: Export results as professional PDF documents
• Bilingual Support: Full support for both English and Arabic

Prerequisites

• Python 3.9 or higher
• OpenAI API key
• (Optional) Anthropic API key for alternative model support

Installation

Option 1: Using Replit

1. Fork this Replit project

2. Set up your environment secrets:

   • Go to the "Secrets" tab in your Replit project
   • Add your OPENAI_API_KEY as a new secret
   • (Optional) Add ANTHROPIC_API_KEY if you want to use Claude models

3. Run the application:

   • Click the "Run" button at the top
   • The application will start at the URL shown in the console

Option 2: Local Installation

1. Clone the repository:

   git clone https://github.com/alanqoudif/Omani-Legal-AI-Assistant.git
   cd omani-legal-ai-assistant
   

2. Install the required packages:

   pip install -r requirements.txt
   

   Required packages include:

   • streamlit
   • openai
   • anthropic (optional)
   • langchain
   • langchain-community
   • numpy
   • faiss-cpu
   • google-generativeai
   • pymupdf
   • reportlab
   • pydub
   • trafilatura
   • arabic-reshaper
   • python-bidi
   • whisper
   • tensorflow
   • tensorflow-hub
   • sentence-transformers

3. Set up environment variables:

   # On Linux/Mac
   export OPENAI_API_KEY="your-openai-api-key"
   
   # On Windows
   set OPENAI_API_KEY=your-openai-api-key
   

4. Run the application:

   streamlit run app.py --server.port 5000
   

Using the Application

1. Process Legal Documents:

   • Click on "Process Legal Documents" to upload PDF files of legal documents
   • The system will process and index the documents for searching

2. Document Q&A:

   • Ask questions about the legal documents
   • Use voice input by clicking the microphone button or uploading an audio file
   • View responses with improved text formatting

3. Article Summarizer:

   • Select a law and article number
   • Get a concise summary of the article

4. Law Comparison:

   • Select two different laws
   • View a detailed comparison highlighting similarities and differences

5. Case Analyzer:

   • Describe a legal case
   • Receive analysis based on relevant laws
   • Download a PDF report

6. Document Creator:

   • Specify details for a legal document
   • Get a professionally formatted document

Language Support

• Toggle between English and Arabic using the language selector
• All components support both languages
• Arabic text is properly handled with bidirectional support

API Keys

This application uses OpenAI's GPT models for text processing. You'll need to provide your own API key:

1. Get an API key from OpenAI
2. Add it to your environment variables as OPENAI_API_KEY

For Claude model support:

1. Get an API key from Anthropic
2. Add it to your environment variables as ANTHROPIC_API_KEY

Troubleshooting

• Application not starting: Ensure all dependencies are installed and API keys are set up
• PDF processing errors: Make sure PDFs are not encrypted and are in a readable format
• Arabic text display issues: Verify that arabic-reshaper and python-bidi are installed correctly
• Voice input not working: Check that audio files are in supported formats (WAV, MP3, OGG)

License

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

Acknowledgments

• This project uses various AI models and libraries for natural language processing
• Special thanks to the Streamlit, Langchain, and OpenAI communities for their tools and documentation