Summary
If you're building live voice or video chat with Gemini, this covers the WebSocket-based Live API end to end. It handles bidirectional audio streaming, video frames, function calling, voice activity detection for interruptions, and the newer native audio features. The guide is opinionated about using gemini-3.1-flash-live-preview and the new google-genai SDKs, with clear warnings about deprecated models. Honestly refreshing to see concrete audio format specs (16-bit PCM at 16kHz input, 24kHz output) and partner integrations like LiveKit if you'd rather use WebRTC. Good for real-time conversational AI where you need low latency and don't want to guess at the config options.
Install to Claude Code
npx -y skills add google-gemini/gemini-skills --skill gemini-live-api-dev --agent claude-codeInstalls into .claude/skills of the current project.
belt cli automatically finds the best tools and skills for your agent. image, video, music, tts...
one prompt install →
Plug Mailtrap into your AI workflow and let it handle the email.
Connect Mailtrap MCP →
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
Make coding agent sessions - Searchable, Shareable, Vendor-neutral & Scored.
Try For Free →
Your agent targets a perfect 10 Code Health score. Deterministic. Every commit.
Try For Free →
Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →Files
SKILL.md
Select a file.
Featured
belt cli automatically finds the best tools and skills for your agent. image, video, music, tts...
one prompt install →Plug Mailtrap into your AI workflow and let it handle the email.
Connect Mailtrap MCP →Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →Make coding agent sessions - Searchable, Shareable, Vendor-neutral & Scored.
Try For Free →Your agent targets a perfect 10 Code Health score. Deterministic. Every commit.
Try For Free →Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →
