For AI agents: a documentation index is available at the root level at /llms.txt and /llms-full.txt. Append /llms.txt to any URL for a page-level index, or .md for the markdown version of any page.
Ask AIPlaygroundLoginFree API Key
HomeAPI ReferenceVoice AgentSpeech-to-TextText-to-SpeechIntelligenceSelf-Hosted Deployments
HomeAPI ReferenceVoice AgentSpeech-to-TextText-to-SpeechIntelligenceSelf-Hosted Deployments
  • Get Started
    • Overview
    • Build a Voice Agent
    • Feature Overview
    • Template Apps
  • Configure
    • Overview
    • STT Models
    • LLM Models
    • TTS Models
    • Media Inputs & Outputs
    • Prompting Voice Agents
    • Multilingual Voice Agents
    • Maintaining Context
    • Reusable Agent Configurations
  • Build
    • Multi-Agent Architecture
  • Connect
  • Controls
      • Overview
      • Settings
      • Update Speak
      • Update Think
      • Update Prompt
      • Inject Agent
      • Inject User
      • Agent Keep Alive
  • Optimize
    • Voice Agent TTS Controls
    • Message Flow
    • Audio & Playback
    • Audio Preprocessing & Barge-In
    • Adaptive Echo Cancellation
  • Resources
    • SDKs
    • UI Components
    • API Reference
LogoLogo
Ask AIPlaygroundLoginFree API Key
On this page
  • List of Client Messages
ControlsInputs: Client Messages

Inputs: Client Messages

Client messages you can send to the server to control the Voice Agent.
Was this page helpful?
Previous

Settings

Send a Settings message to configure the voice agent's behavior, audio formats, and provider configurations before starting the conversation.
Next
Built with

Client Messages are JSON-formatted commands that the client sends to the Deepgram server over the WebSocket connection during a voice agent interaction. These messages allow the client to control various aspects of the conversation, configure the agent, and provide necessary information.

List of Client Messages

  • Settings: Initializes the voice agent, sets up audio transmission formats, and optionally provides conversation history context before any voice data is exchanged.
  • Update Speak: Enables changing the Speak model during the conversation.
  • Update Think: Replaces the entire Think provider configuration, including the model, prompt, endpoint, and functions.
  • Inject Agent Message: Triggers an immediate statement from the agent.
  • Inject User Message: Sends a text-based message to the agent as if the user had spoken it.
  • Update Prompt: Updates the system prompt of the agent during the conversation.
  • Function Call Response: Sends the result of a function call back to the server.
  • Agent Keep Alive: Maintains the connection to prevent timeouts.

For more detailed information on the format and usage of each message, refer to the individual documentation pages for each message.