Public preview: Voice-native agents in Microsoft Foundry

2026-03-16 · Source: Microsoft Foundry Blog articles · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering, Cloud Computing & IT Infrastructure · Depth: Intermediate, medium

Summary

Microsoft has introduced the public preview of Voice Live with the new Foundry Agent Service, enabling developers to build real-time, speech-to-speech AI agents on Azure more easily. This integration unifies agent orchestration and real-time voice interaction into a single developer experience, streamlining the creation of production-ready voice-native agents. The Foundry Agent Service, now generally available, features a redesigned API and runtime with built-in evaluators for quality, groundedness, safety, and risk. Voice Live is a real-time API offering premium speech capabilities, consolidating functionalities like speech recognition, text-to-speech, natural turn detection, interruption handling, and avatar integration into one API call. It supports over 140 locales with more than 700 voices, including 40+ neural HD conversational voices, and provides advanced conversational enhancements like noise suppression and echo cancellation. This service aims to accelerate development and reduce engineering complexity for voice-enabled agents.

Key takeaway

For AI Architects and CTOs evaluating platforms for conversational AI, the integration of Microsoft's Foundry Agent Service and Voice Live API significantly reduces development complexity and time-to-value for real-time, speech-to-speech agents. You should explore this unified service to build enterprise-grade voice agents that handle planning, reasoning, and tool execution, enabling natural, low-latency interactions and accelerating deployment of intelligent applications.

Key insights

Microsoft's new Foundry Agent Service and Voice Live API simplify building real-time, speech-to-speech AI agents on Azure.

Principles

• Unify real-time voice and agent logic.
• Prioritize natural, low-latency conversations.
• Integrate evaluation throughout the AI lifecycle.

Method

The system processes user speech via Voice Live API, converts it to conversational input, the Agent Service reasons and executes tools, then synthesizes and streams the response as natural speech.

In practice

• Use Foundry Agent Service for real-time self-service support.
• Embed voice agents into workflows for operational tasks.
• Develop hands-free in-car voice assistants.

Topics

• Microsoft Foundry Agent Service
• Voice Live API
• Real-time Conversational AI
• Speech-to-Speech Agents
• Azure AI

Best for: AI Architect, CTO, VP of Engineering/Data, AI Engineer, Machine Learning Engineer, MLOps Engineer

Related on AIssential

• New Azure Open AI models bring fast, expressive, and real‑time AI experiences in Microsoft Foundry — New Azure OpenAI models prioritize continuity and reliability for complex, real-time AI applications and long-running engineering tasks.
• Scalable voice agent design with Amazon Nova Sonic: multi-agent, tools, and session segmentation — Multi-agent architectures and session segmentation are key to building scalable, low-latency voice agents.
• Teaching AI to Remember: Exploring Memory Store in Microsoft Foundry — Microsoft Foundry's Memory Store provides managed, user-scoped long-term memory for AI agents, enabling personalized, continuous interactions.
• Azure AI Foundry: The Architect’s Blueprint for Building Enterprise AI at Scale — Azure AI Foundry is a platform of ten capabilities requiring deliberate architectural choices and external governance layers for enterprise AI.
• Deploy voice agents with Pipecat and Amazon Bedrock AgentCore Runtime – Part 1 — Achieving natural, low-latency voice AI requires robust streaming architectures and scalable runtime environments.
• Introducing Fireworks AI on Microsoft Foundry: Bringing high performance, low latency open model inference to Azure — Fireworks AI on Microsoft Foundry provides high-performance, enterprise-grade open model inference within Azure's unified AI platform.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Microsoft Foundry Blog articles.