Speech-to-speech realtime
Streams caller audio and returns spoken responses over a persistent WebSocket session.
AI Models
Grok Voice
xAI's full-duplex Grok Voice model for live phone agents. Stream call audio in and play spoken replies back in the same session.
Speech-to-speechRealtimeTool callingWebSocketFull-duplex
Where it fits
Grok Voice in your voice stack.
- TelephonyPhone networkInbound call
- AgentDuetCall + audioAnswer, stream
- Speech-to-speechGrok VoiceListens, call functions, replies
- ToolsActionsCRM, calendar, APIs
Overview
What it contributes to the stack.
Grok Voice is xAI's full-duplex speech model for live agents. It streams spoken replies, supports natural turn-taking, and can use tools during the call.
With AgentDuet, the phone connection stays on AgentDuet. Your application streams caller audio into Grok Voice and plays speech back to the caller.
For setup, credentials, and API details, use the Documentation link above.
Capabilities
The technical characteristics that matter when deciding whether this provider belongs in your application.
Tool calling
Supports function tools and search-style capabilities during the live conversation.
Server turn detection
Can use provider-managed voice activity detection for natural turn-taking.
Telephony-ready audio
Works with call audio streams that map cleanly to AgentDuet sessions.
Common use cases
Practical scenarios where the provider's role is clear and the surrounding systems remain under application control.
Customer support
Handle spoken requests while tools retrieve account or product facts.
Sales conversations
Qualify inbound callers and capture structured next steps.
Always-on assistants
Keep a responsive voice agent available across phone sessions.
Related integrations
Compare nearby choices.
These providers occupy a similar role in the stack. Compare their strengths, operating model, and surrounding services.
Browse every integration
Build a voice agent.
AgentDuet handles the phone and messaging boundary. Your model runs the conversation.