OMANI-Therapist-Voice

A comprehensive mental health support chatbot that communicates exclusively through voice in Omani Arabic dialect, providing culturally sensitive, therapeutic-grade conversations with real-time speech processing capabilities.

Project Structure

omani-therapist-voice/
├── frontend/                 # Next.js React application
│   ├── src/
│   │   ├── components/    # React components
│   │   ├── stores/        # Custom React hooks for store
│   │   ├── utils/         # Utility functions
│   │   └── app/           # React Page Layout
│   └── package.json
├── backend/                 # FastAPI Python application
│   ├── app/
│   │   ├── api/           # API routes
│   │   ├── core/          # Core functionality
│   │   ├── models/        # Data models
│   │   ├── services/      # Business logic
│   │   └── utils/         # Utility functions
│   ├── requirements.txt
│   └── main.py
└── docs/                   # Documentation

Features

• Real-time Voice Processing: Audio capture, STT, and TTS in Omani Arabic
• Dual-Model AI: GPT-4o with Google Gemini fallback/validation
• Cultural Sensitivity: Gulf-specific mental health terminology
• Therapeutic Techniques: CBT adaptation and crisis intervention
• Safety Protocols: Suicide risk assessment and escalation
• HIPAA Compliance: Secure data handling and privacy protection

Quick Start

1. Clone the repository
2. Set up environment variables

Architecture

The system uses a microservices architecture with:

• Frontend: Next.js with WebRTC for audio capture
• Backend: FastAPI with async processing
• AI Models: OpenAI GPT-4o,Google Gemini
• Speech Services: OpenAI Services for STT/TTS
• Database: PostgreSQL for session storage

Ethical Considerations

• Transparent AI disclosure
• Data privacy and anonymization
• Professional supervision integration
• Cultural competency validation