Bidi-streaming (live) in ADK¶
ADK 中的双向流式传输(live)¶
Bidirectional (Bidi) streaming (live) in ADK adds the low-latency bidirectional voice and video interaction capability of Gemini Live API to AI agents.
ADK 中的双向流式传输(live)模式将Gemini Live API的低延迟双向语音和视频交互功能添加到 AI 智能体。
With bidi-streaming, or live, mode, you can provide end users with the experience of natural, human-like voice conversations, including the ability for the user to interrupt the agent's responses with voice commands. Agents with streaming can process text, audio, and video inputs, and they can provide text and audio output.
使用双向流式传输或 live 模式,您可以为最终用户提供自然、类似人类的语音对话体验,包括用户使用语音命令中断智能体响应的能力。具有流式传输功能的智能体可以处理文本、音频和视频输入,并且它们可以提供文本和音频输出。
-
Quickstart (Bidi-streaming)
In this quickstart, you'll build a simple agent and use streaming in ADK to implement low-latency and bidirectional voice and video communication.
在此快速入门中,您将构建一个简单的智能体,并在 ADK 中使用流式传输来实现低延迟和双向语音和视频通信。
-
Bidi-streaming Demo Application
A production-ready reference implementation showcasing ADK bidirectional streaming with multimodal support (text, audio, image). This FastAPI-based demo demonstrates real-time WebSocket communication, automatic transcription, tool calling with Google Search, and complete streaming lifecycle management. This demo is extensively referenced throughout the development guide series.
一个生产就绪的参考实现,展示了具有多模态支持(文本、音频、图像)的 ADK 双向流式传输。这个基于 FastAPI 的演示展示了实时 WebSocket 通信、自动转录、使用 Google Search 的工具调用以及完整的流式传输生命周期管理。此演示在开发指南系列中被广泛引用。
-
Blog post: ADK Bidi-streaming Visual Guide
A visual guide to real-time multimodal AI agent development with ADK Bidi-streaming. This article provides intuitive diagrams and illustrations to help you understand how Bidi-streaming works and how to build interactive AI agents.
一份关于使用 ADK 双向流式传输进行实时多模态 AI 智能体开发的视觉指南。本文提供了直观的图表和插图,帮助您理解双向流式传输的工作原理以及如何构建交互式 AI 智能体。
-
Bidi-streaming development guide series
A series of articles for diving deeper into Bidi-streaming development with ADK. You can learn basic concepts and use cases, core API, and end-to-end application design.
一系列文章,深入探讨使用 ADK 进行双向流式传输开发。您可以学习基本概念和用例、核心 API 以及端到端应用程序设计。
- Part 1: Introduction to ADK Bidi-streaming - Fundamentals of Bidi-streaming, Live API technology, ADK architecture components, and complete application lifecycle with FastAPI examples 第 1 部分:ADK 双向流式传输简介 - 双向流式传输的基础知识、Live API 技术、ADK 架构组件以及使用 FastAPI 示例的完整应用程序生命周期
- Part 2: Sending messages with LiveRequestQueue - Upstream message flow, sending text/audio/video, activity signals, and concurrency patterns 第 2 部分:使用 LiveRequestQueue 发送消息 - 上游消息流、发送文本/音频/视频、活动信号和并发模式
- Part 3: Event handling with run_live() - Processing events, handling text/audio/transcriptions, automatic tool execution, and multi-agent workflows 第 3 部分:使用 run_live() 处理事件 - 处理事件、处理文本/音频/转录、自动工具执行和多智能体工作流
- Part 4: Understanding RunConfig - Response modalities, streaming modes, session management, session resumption, context window compression, and quota management 第 4 部分:理解 RunConfig - 响应模态、流式传输模式、会话管理、会话恢复、上下文窗口压缩和配额管理
- Part 5: How to Use Audio, Image and Video - Audio specifications, model architectures, audio transcription, voice activity detection, and proactive/affective dialog features 第 5 部分:如何使用音频、图像和视频 - 音频规范、模型架构、音频转录、语音活动检测和主动/情感对话功能
-
Streaming Tools
Streaming tools allow tools (functions) to stream intermediate results back to agents and agents can respond to those intermediate results. For example, we can use streaming tools to monitor changes of a stock price and have the agent react to it. Another example is we can have the agent monitor a video stream, and when there are changes in the video stream, the agent can report the changes.
流式传输工具允许工具(函数)将中间结果流式传输回智能体,智能体可以响应这些中间结果。例如,我们可以使用流式传输工具来监控股票价格的变化并让智能体对此做出反应。另一个例子是我们可以让智能体监控视频流,当视频流中有变化时,智能体可以报告这些变化。
-
Blog post: Google ADK + Vertex AI Live API
This article shows how to use Bidi-streaming (live) in ADK for real-time audio/video streaming. It offers a Python server example using LiveRequestQueue to build custom, interactive AI agents.
本文展示了如何在 ADK 中使用双向流式传输(live)进行实时音频/视频流式传输。它提供了一个使用 LiveRequestQueue 构建、自定义交互式 AI 智能体的 Python 服务器示例。
-
Blog post: Supercharge ADK Development with Claude Code Skills
This article demonstrates how to use Claude Code Skills to accelerate ADK development, with an example of building a Bidi-streaming chat app. Learn how to leverage AI-powered coding assistance to build better agents faster.
本文演示了如何使用 Claude Code Skills 来加速 ADK 开发,并附有一个构建双向流式传输聊天应用程序的示例。了解如何利用 AI 驱动的编码辅助来更快地构建更好的智能体。