Build a voice travel concierge with Amazon Bedrock AgentCore, Managed Knowledge Base and Nova Sonic
Machine Learning Blog
This article demonstrates how to build a voice travel concierge for airline applications using managed AWS services, enabling travelers to manage bookings through natural voice commands.
- Deploy voice AI concierge using Amazon Bedrock AgentCore runtime with Amazon Nova 2.5 Sonic for real-time speech processing
- Connect agent to backend services via Model Context Protocol (MCP) through AgentCore Gateway without tight coupling
- Answer airline policy questions using Amazon Bedrock Knowledge Bases with retrieval-augmented generation
- Authenticate users with Amazon Cognito and secure WebSocket connections using SigV4 signatures
- Handle seat changes, itinerary lookups, meal preferences, and escalations to live agents with confirmation workflows
- Deploy entire solution with AWS CDK using DynamoDB, Lambda, API Gateway, and Amazon S3
- Leverage asynchronous tool calling and latency masking for natural, responsive conversations
The layered architecture separates frontend, AI agent, and backend services for independent scaling and development, with pay-per-use pricing and automatic scaling for traffic spikes.
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.
Related articles
2026
2025
2026
2026
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.