Can I use this SDK in a browser environment?

Yes, it is compatible with modern browsers (Chrome, Firefox, Safari, Edge) when used with a bundler like Vite or Webpack.

What models are supported by this Azure Voice skill?

The skill supports high-performance models including gpt-4o-realtime-preview, gpt-4o-mini-realtime-preview, and phi4-mm-realtime.

How does the skill handle authentication?

It supports both Microsoft Entra ID (via @azure/identity) and standard API Key authentication using AzureKeyCredential.

What is Semantic VAD in this context?

Semantic VAD is an advanced Voice Activity Detection system that uses AI to better understand when a user has finished speaking, leading to more natural conversations.

Azure AI Voice Live (TypeScript)

Name: Azure AI Voice Live (TypeScript)
Author: microsoft

bymicrosoft

•

1,777

•

云基础设施

Builds real-time voice AI applications with bidirectional WebSocket communication using the Azure AI Voice Live SDK.

This skill enables developers to integrate low-latency, real-time voice capabilities into JavaScript and TypeScript applications. It provides structured guidance for managing VoiceLiveClient sessions, handling bidirectional audio streams, and implementing advanced features like Azure Semantic VAD (Voice Activity Detection) and function calling. Whether building for Node.js or the browser, this skill helps orchestrate complex voice interactions using state-of-the-art models like GPT-4o Realtime, making it ideal for creating sophisticated AI voice assistants and speech-to-speech applications.

主要功能

01Bidirectional WebSocket communication for real-time, low-latency audio streaming.

021,777 GitHub stars

03Support for GPT-4o Realtime and Phi-4 multimodal models.

04Advanced Turn Detection including Server VAD and Azure Semantic VAD.

05Integrated function calling support for interactive voice-enabled tools.

06Comprehensive audio format support including PCM16 and G.711 for telephony.

使用场景

01Creating natural-sounding AI voice assistants with intelligent turn-taking.

02Implementing voice-activated customer support interfaces in web and Node.js applications.

03Building real-time speech-to-speech translation tools for global communication.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add microsoft/skills azure-ai-voicelive-ts

For use in Claude.ai and ChatGPT

Download Skill