Uplift

AI therapist app with real-time chat and voice streaming. Flutter frontend, FastAPI backend, OpenAI for conversation and TTS.

How It Works

User sends text or voice input from the Flutter app. The app connects to the backend over WebSockets for streaming responses. The backend calls OpenAI for conversation and text-to-speech, streaming audio back in real time.

User input (text/voice) → Flutter app → FastAPI backend → OpenAI → streamed response → app

Chat: ws://localhost:8000/ws/chat (JSON messages, streamed responses)

Voice: ws://localhost:8000/voice/ws/tts (streamed audio output)

Features

• Text and voice conversations with context-aware responses
• Real-time streaming for both chat and TTS audio
• Session management with conversation history
• Cross-platform: Android (primary), iOS, Windows, macOS

Architecture

Layer	Technology
Frontend	Flutter 3.0+, BLoC state management
Backend	FastAPI (Python 3.9+)
AI	OpenAI (chat completions + TTS)
Transport	REST + WebSocket

Quick Start

Backend

cd backend
python -m venv env && source env/bin/activate
pip install -r requirements.txt
cp .env.example .env  # add your OPENAI_API_KEY
uvicorn app.main:app --reload

Frontend

cd frontend
flutter pub get
flutter run

Update lib/config.dart with your backend URL if not running on localhost.

Testing

# Backend
pytest tests/

# Mock mode (no API key needed)
# Set USE_MOCK=true in .env

License

MIT