gemini-live-api-dev

Stream audio, video, and text via the Gemini Live API over WebSockets.

2|Updated Jan 29, 2026
One-click install
npx skills add https://github.com/Tai-ch0802/skills-bundle --skill gemini-live-api-dev-tai-ch0802
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: gemini-live-api-dev
Source: https://github.com/Tai-ch0802/skills-bundle/tree/main/i18n/zh-TW/gemini-live-api-dev
Command: npx skills add https://github.com/Tai-ch0802/skills-bundle --skill gemini-live-api-dev-tai-ch0802

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill streamlines the development of real-time, low-latency applications that require interactive voice and video communication with the Gemini API, handling complex streaming protocols and features.

Core Features & Use Cases

  • Bi-directional Streaming: Enables real-time audio, video, and text conversations.
  • Advanced Audio Features: Supports Voice Activity Detection (VAD), native audio for emotional dialogue, and active audio.
  • Function Calling & Grounding: Integrates tool usage and Google Search for context-aware responses.
  • Session Management: Handles context compression, session recovery, and authentication.
  • Use Case: Build a live customer support chatbot that can understand and respond to users via voice in real-time, complete with natural-sounding speech and the ability to trigger actions based on the conversation.

Quick Start

Connect to the Live API using the Python SDK and send a text message to initiate a conversation.

Frequently Asked Questions about gemini-live-api-dev

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I build a real-time streaming app with the Gemini Live API?

The Gemini Live API supports real-time, bi-directional audio, video, and text streaming via WebSockets. It provides Voice Activity Detection, native audio for emotional dialogue, and active audio features for interactive communication applications.

Can I use WebSockets for real-time video and audio streaming with Gemini?

Yes, you can use WebSockets for real-time video and audio streaming with Gemini. The Live API facilitates bi-directional streaming, allowing you to send and receive media streams directly through the google-genai and @google/genai SDKs.

Does the Gemini Live API support voice activity detection and native audio?

Yes, the Gemini Live API supports Voice Activity Detection (VAD) and native audio for emotional dialogue. It also includes active audio features to facilitate natural-sounding, real-time voice conversations.

How do I integrate function calling into a real-time audio streaming session?

You integrate function calling into a real-time audio streaming session using the Gemini Live API. This enables context-aware responses by triggering external tools and Google Search during live bi-directional conversations.

What SDKs are supported for Gemini Live API real-time streaming development?

The Gemini Live API supports real-time streaming development using the Python (google-genai) and JavaScript/TypeScript (@google/genai) SDKs. Both provide interfaces for session management, WebSocket connections, and ephemeral token authentication.

How does the Gemini API handle session recovery and context compression?

The Gemini API manages session recovery and context compression through built-in session management features. This allows real-time streaming applications to maintain continuous conversations while handling authentication and ephemeral tokens.