Agents are the core of the voice experience. They are powered by an LLM, and can be configured to fit any use case. In this section, you will learn how to configure your agents to create a voice experience that is tailored to your needs.

​
Configuration

This section is divided into four main parts, each of which is responsible for a different aspect of the agent’s configuration.

Voice Config

This is the configuration for the voice that the agent will use to speak to the user.

"voice": {
  "provider": "elevenlabs",
  "model": "adam"
}

Language Model Config

This is how the agent can understand the user’s intent and respond accordingly.

"language_model": {
  "provider": "openai",
  "model": "gpt-4-preview",
  "tools": []
}

Transcriber Config

This is how the agent can hear and understand what the user is saying in the conversation.

"transcriber": {
  "provider": "deepgram",
  "model": "nova-2"
}

Webhook Config

This is how you can start integrating your agent with your own backend logic.

"webhooks": [
  {
    "events": ["speech:started", "message:generated", "function_call:triggered"],
    "url": "your-backend-url",
    "method": "POST",
    "headers": {
      "Content-Type": "application/json",
      "Authorization": "your-auth-schema"
    }
  }
]
Our SDKsConversations