To enqueue asynchronous audio rendering assignments directly onto the server generation channels, execute a speak_request packet command:

{
  "type": "speak_request",
  "payload": {
    "text": "Speech pipeline automated transmission message."
  }
}

Speak Parameter Specifications

JSON KeyTypeSpecification Description
typestringMust be 'speak_request' or 'tts_request' interchangeably to activate text rendering pipeline workflows.
payload.textstringThe raw message text phrase sent to process into synthesized voice data array frames.