Building a voice-driven AWS assistant with Amazon Nova Sonic

Machine Learning Blog

This article demonstrates building a voice-driven AWS operations assistant using Amazon Nova Sonic and Strands Agents for multi-agent orchestration.

Uses Amazon Nova Sonic for speech processing and natural language understanding
Multi-agent architecture with supervisor agent routing to specialized EC2, SSM, and Backup agents
Enables voice commands like "List EC2 instances" or "Check backup status" with intelligent responses
Backend uses Python 3.12+ with Strands Agents; frontend uses React with AWS Cloudscape
Real-time bidirectional voice processing via WebSocket server
Responses optimized for voice delivery, limited to 800 characters
Supports EC2 management, SSM operations, backup monitoring, and extensible to other services
Production deployment recommended using Amazon Bedrock AgentCore Runtime for enterprise security
Complete source code available in GitHub repository

This solution transforms AWS infrastructure management through conversational AI, making cloud operations more accessible to technical and non-technical users alike.

Go to article

The AWS News Feed is currently looking for gold sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.

Nov 26
2025

Building AI-Powered Voice Applications: Amazon Nova Sonic Telephony Integration Guide

Oct 21
2025

Building a multi-agent voice assistant with Amazon Nova Sonic and Amazon Bedrock AgentCore

Feb 10
2026

Building real-time voice assistants with Amazon Nova Sonic compared to cascading architectures

Jul 18
2025

Deploy a full stack voice AI agent with Amazon Nova Sonic

The AWS News Feed is currently looking for silver sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.

Building a voice-driven AWS assistant with Amazon Nova Sonic

Related articles