LiveKit Agents开源框架，实时语音、视频 AI 代理

很输号、

这份文档是组织成几大块的： Introduction：开始这里去理解 LiveKit 的核心概念和做好准备。 Build Agents：学习使用 LiveKit Agents 框架构建 AI 代理。 Agent Frontends：为 AI 代理构建 web、移动和硬件界面。 Telephony：将 AI 代理与电话网络和传统通信系统连接。 WebRTC Transport：深入 WebRTCConcepts 和底层传输细节。 Manage & Deploy：部署和管理 LiveKit 代理和基础设施，并学习测试、评估和观察代理性能。 Reference：API 参考、 SDK 文档 和组件库。 使用侧边栏导航来探索每个节中的主题。 每个页面都包括代码示例、指南和与相关概念相关的链接。首先阅读Understanding LiveKit概要以了解核心概念，然后按照您的用例跟随指南。

Asked by Fumiko Sato · Nov 19, 2025

LiveKit是什么样子的？

LiveKit的架构由多个关键组件组成，它们协同工作

Asked by Yuki Kobayashi · Oct 31, 2025

我可以用LiveKitAgents构建什么？

LiveKit支持各种应用：音频视频文本交互的多模态AI助手和Avatar。视频会议：团队的任何大小私有团队聚会。互动直播：实时与观众交流。客服：可观察、灵活的网页、手机、电话的支持选项。医疗：人机协同的HIPAA符合的远程医疗。机器人：将实时视频、强大的AI模型集成到现实世界的设备中。LiveKit 为运行生产级AI体验所需的实时基础（低延迟、高可扩展性、灵活工具）提供支持

Asked by Ravi Chandrasekaran · Sep 20, 2025

Why use LiveKit?

LiveKit differentiates itself through several key advantages: Build faster with high-level abstractions: Use the LiveKit Agents framework to quickly build production-ready AI agents with built-in support for speech processing, turn-taking, multimodal events, and LLM integration. When you need custom behavior, access lower-level WebRTC primitives for complete control. Write once, deploy everywhere: Both human clients and AI agents use the same SDKs and APIs, so you can write agent logic once and deploy it across Web, iOS, Android, Flutter, Unity, and backend environments. Agents and clients interact seamlessly regardless of platform. Focus on building, not infrastructure: LiveKit handles the operational complexity of WebRTC so developers can focus on building agents. Choose between fully managed LiveKit Cloud or self-hosted deployment — both offer identical APIs and core capabilities. Connect to any system: Extend LiveKit with egress, ingress, telephony, and server APIs to build end-to-end workflows that span web, mobile, phone networks, and physical devices.

Asked by Chioma Nwosu · Sep 6, 2025

What is LiveKit?

LiveKit is an open source framework and cloud platform for building voice, video, and physical AI agents. It provides the tools you need to build agents that interact with users in realtime over audio, video, and data streams. Agents run on the LiveKit server, which supplies the low-latency infrastructure (including transport, routing, synchronization, and session management) built on a production-grade WebRTC stack. This architecture enables reliable and performant agent workloads.

Asked by Greta Nowak · Aug 9, 2025