Your smart voice-powered butler for research, scheduling & smart updates via calls & messages.

• Research Management: Initiate and retrieve research on any topic
• Smart Scheduling: Event planning with contextual reminders
• Task Management: Create and track tasks with customizable priorities
• Contact Management: Store and manage contact information
• SMS & Call Integration: Automated notifications and voice interactions
• Timezone Support: Global timezone handling for events and reminders

• Python 3.11+
• PostgreSQL (via Supabase)
• Twilio Account (for SMS/calls)

1. Clone the repository:

   git clone <repository-url>
   cd jarvoice-api

2. Create and activate virtual environment:

   python -m venv venv
   source venv/bin/activate  # Unix/macOS
   # or
   .\venv\Scripts\activate  # Windows

3. Install dependencies:

   pip install -r requirements.txt

4. Set up environment variables:

   cp .env.example .env.local

   Required environment variables:

   SUPABASE_URL=your_supabase_url
   SUPABASE_ANON_KEY=your_supabase_key
   TWILIO_ACCOUNT_SID=your_twilio_sid
   TWILIO_AUTH_TOKEN=your_twilio_token
   

uvicorn main:app --reload

The API will be available at http://localhost:8000

Once running, visit:

• Swagger UI: http://localhost:8000/docs
• ReDoc: http://localhost:8000/redoc

• POST /users: Create new user
• GET /users/{user_id}: Retrieve user details
• POST /tasks: Create new task
• POST /events: Create calendar event
• POST /contacts: Add new contact
• POST /1/process: Process tool calls for AI interactions

jarvoice-api/
├── main.py              # FastAPI application and routes
├── base_models.py       # Pydantic models
├── tool_functions.py    # AI tool implementations
├── scheduler.py         # Task scheduling logic
├── outbound_caller.py   # Voice call handling
├── twilio_sms.py        # SMS functionality
└── tool_registry.py     # Tool function registry

Tools in Vapi are implemented through server URLs and function calls, enabling your assistant to interact with external services.

Server URLs allow real-time communication between Vapi and your application. They enable you to:

• Receive conversation events (status updates, transcripts, function calls)
• Get dynamic assistant configurations
• Receive end-of-call reports

To set up a server URL:

1. Configure the URL in your assistant settings.
2. Implement an endpoint on your server to handle POST requests from Vapi.
3. Process the received data and respond as needed.

Function calls allow your assistant to trigger specific actions during a conversation. You can define custom functions that your assistant can invoke, such as:

• Forwarding calls
• Hanging up
• Accessing external databases or APIs

For testing server URLs in a local development environment, you can use ngrok:

1. Install ngrok.
2. Run your local server.
3. Use ngrok to create a public URL for your local server.
4. Set the ngrok URL as your server URL in Vapi.

You can plug in your own Language Model by configuring a custom LLM URL, allowing you to use specialized or proprietary models for your assistant.

To integrate with external services (e.g., specific TTS, LLM, or STT vendors), add your API keys in the Vapi dashboard.

Assistants can be customized with various components:

• Transcriber: For speech-to-text conversion
• Model: The language model powering the assistant's responses
• Voice: The text-to-speech voice for the assistant

By leveraging these tools and configurations, you can create highly customized and powerful voice AI assistants using Vapi.

1. Define new models in base_models.py
2. Implement business logic in appropriate modules
3. Add new endpoints in main.py
4. Register new tool functions in tool_registry.py

# Run tests
pytest

# Test specific functionality
python -m pytest tests/test_specific.py

Key packages:

• fastapi: Web framework
• pydantic: Data validation
• uvicorn: ASGI server
• supabase: Database client
• twilio: SMS/Voice capabilities
• vapi: Voice API integration
• apscheduler: Task scheduling

• CORS middleware configured
• Environment-based configuration
• Supabase authentication
• E.164 phone number validation

1. Fork the repository
2. Create feature branch (git checkout -b feature/amazing-feature)
3. Commit changes (git commit -m 'Add amazing feature')
4. Push to branch (git push origin feature/amazing-feature)
5. Open Pull Request

MIT License

For support or queries, please open an issue in the repository.

Let me know if you'd like further adjustments!