voice_agent

Add an AI participant that listens to room audio and responds with synthesized speech.

Providers

Provider	provider value	Docs
OpenAI Realtime	openai_realtime	Speech-to-speech over WebSocket
Google Gemini Live	gemini_live	Full-duplex audio
Cloudflare Workers AI	cloudflare	Your own Cloudflare voice agent Worker

Example

{
  "name": "bot",
  "type": "voice_agent",
  "inputs": ["mix_audio"],
  "config": {
    "provider": "openai_realtime",
    "instructions": "You are a helpful live show co-host.",
    "greeting": "Hello everyone!",
    "config": {
      "apiKey": "<openai-api-key>",
      "voice": "alloy"
    }
  }
}

Config

Field	Description
provider	Provider id (see table)
instructions	System prompt (max 8000 chars)
greeting	Optional first utterance (max 2000 chars)
config	Provider-specific credentials and options

Pricing

$0.025/min while the agent is actively processing audio. Node pricing.