Skip to main content
Realtime agents listen and respond with voice as a conversation happens. Enable Realtime Voice when your agent needs to handle live conversations through supported voice-enabled channels. The agent still uses its assigned prompt, model, knowledge bases, and integrations. When configuring a realtime agent, you can choose the voice model and synthesized voice, decide whether callers can interrupt the agent, and select the sound played while the agent completes tool calls. Phone-based agents can also use Call Recording and Call-Start Execution — see the pages below for each.
1

Enable Realtime Voice

Open the agent’s settings and toggle Realtime Voice to Enabled.
2

Choose a Model and Voice

Select the realtime model that will power the conversation, then preview and choose the voice that best fits the agent’s purpose.
3

Configure Conversation Behavior

Choose whether callers can interrupt the agent while it is speaking. You can also select the processing sound, volume, and speed used while the agent works through tool calls.
4

Configure Phone Call Features

For Phone, SIP, and Five9 channels, enable Call Recording or Call-Start Execution if the agent needs those features.
5

Test the Agent

Test the opening, interruption behavior, tool calls, and fallback responses before assigning the agent to a live channel.

​
Advanced Phone Features

Call Recording

Review both sides of calls handled by a realtime agent, with a required recording disclaimer.

Call-Start Execution

Look up caller information or run a startup action as a phone call begins.

​
General Best Practices

  • Write prompts for speech. Use short sentences, pronounceable terms, and clear questions.
  • Add instructions for how to recover when it cannot hear or understand the caller.
  • Keep spoken menus short and state transfer keypad options explicitly.
  • Test calling with background noise, interruptions, silence, and unexpected answers.