Voice AI

Voice AI is a full-stack conversational assistant with speech-to-text, text generation, and text-to-speech support.

What’s Included

• FastAPI backend for chat, transcription, and voice chat flows
• Groq-powered LLM responses
• edge_tts-based speech generation
• CORS configuration for local frontend development

Project Structure

• backend/ - FastAPI application, services, and API routes
• frontend/ - frontend workspace placeholder for the client application

Quick Start Backend

1. Copy the environment template and fill in your keys:

cp .env.example backend/.env

2. Create or activate the backend virtual environment and install dependencies:

cd backend
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt

3. Start the backend API:

uvicorn app.main:app --reload

If you prefer running from the repository root, use:

./backend/venv/bin/uvicorn app.main:app --reload --app-dir backend

Environment Variables for Backend

The backend reads these values from .env:

• GROQ_API_KEY - required for chat and transcription features
• CORS_ORIGINS - comma-separated list of allowed frontend origins
• LOG_LEVEL - application log level, defaults to INFO

Notes

• The backend currently uses an in-memory conversation store.
• Audio uploads accept .mp3, .wav, .m4a, .mp4, and .flac.
• The generated TTS files are temporary and removed after response delivery.