Speech-to-speech streaming
Processes incoming speech and emits spoken responses over an open stream.
AI Models
Amazon Nova 2 Sonic
Amazon's speech-to-speech model on Bedrock for live phone agents. Stream call audio in and play spoken replies back in the same session.
Speech-to-speechAWSBedrockBarge-inTool use
Where it fits
Amazon Nova 2 Sonic in your voice stack.
- TelephonyPhone networkInbound call
- AgentDuetCall + audioAnswer, stream
- Speech-to-speechNova 2 SonicListens, call functions, replies
- ToolsActionsCRM, calendar, APIs
Overview
What it contributes to the stack.
Amazon Nova 2 Sonic is a speech-to-speech model on AWS Bedrock for live voice agents. It streams spoken replies, supports barge-in, and can request tools during the conversation.
With AgentDuet, telephony stays on AgentDuet while your application adapts call audio to Bedrock's bidirectional stream and returns speech 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.
Barge-in
Signals interruptions so the client can stop stale audio playback.
Tool use
Requests application functions for lookups, updates, and other actions.
Conversation context
Maintains multi-turn context during an active model session.
Common use cases
Practical scenarios where the provider's role is clear and the surrounding systems remain under application control.
Account servicing
Retrieve approved account details and complete authenticated service requests.
Delivery support
Check shipment state, capture instructions, and report exceptions.
Internal service desk
Diagnose routine issues and create or update support tickets.
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.