Vapi-Live-Call-Control

This Node.js project allows you to initiate calls through a voice assistant API, capture audio from live calls via WebSocket, save the audio as a WAV file, and inject custom messages into the ongoing call. This project is ideal for applications like call monitoring or voice assistants that handle customer support.

Features

• Initiate a Call: Start a call using an assistant and retrieve a listen URL and control URL.
• Capture Audio: Stream audio in real-time and save it as a WAV file.
• Inject Custom Messages: Dynamically inject a custom message during a live call.

Setup

Prerequisites

• Node.js installed on your machine.
• A valid API key from the voice assistant service (VAPI).
• An Assistant ID, Phone Number ID, and Customer Phone Number from the API.

Installation

1. Clone the repository or download the project files.

   git clone https://github.com/HamChowderr/Vapi-Live-Call-Control
   cd vapi-live-call-control

2. Install the required Node.js packages:

   npm install

3. Create a .env file in the root of the project and add your API key:

   VAPI_API_KEY=your_api_key_here

Configuration

Replace the following values in the index.js file:

• Assistant ID: 'AssistantUUID'
• Customer Phone Number: '+13231234567'
• Phone Number ID: 'PhoneUUID'

These values represent your unique API identifiers and customer phone number. Be sure to replace them with your actual values if you are running this in your own environment.

Usage

1. Initiate the Call: Run the script:

   node index.js

2. Answer the Call: After initiating the call, you will be prompted to press ENTER once the call is answered to start capturing the audio.

3. Inject a Custom Message: After answering the call, you can enter a custom message for the assistant to say during the call.

4. Capture and Save Audio: The audio stream will be saved as a WAV file in the root directory with a unique timestamp-based filename.

Project Structure

voice-assistant-call-listener/
│
├── index.js       # Main application logic
├── .env           # Environment file with the API key (not included in the repository)
├── package.json   # Node.js dependencies and scripts
└── README.md      # Project documentation (this file)

Dependencies

• ws - WebSocket library for handling the audio stream.
• axios - HTTP client for making API requests.
• dotenv - Loads environment variables from a .env file.
• readline - Interface for reading user input.

Troubleshooting

• WebSocket Connection Errors: Ensure that the assistant is correctly initiated and the phone call is active before starting the WebSocket connection.
• API Key Issues: Make sure your .env file is correctly configured and your VAPI_API_KEY is valid.