Absolutely! LiveKit warmly supports video and voice agents through a dedicated framework designed to help you build amazing multimodal AI assistants. This means you can have real-time, programmable participants that gracefully handle both audio and video input. To get video input working for your agent, you simply need to set

video_enabled=True

in the

RoomInputOptions

. Video frames are thoughtfully processed using a sampling mechanism, which you can customize to fit different frame rates. On the voice side, the pipeline includes helpful features like turn detection, voice activity detection (VAD), and interruption handling to make interactions smooth and natural. We also offer client-side templates for popular platforms like Swift, Flutter, and React, all supporting video and voice input. Plus, there are complete examples ready for you to explore, including the Gemini video agent and some exciting vision demos to get you started with confidence. If you’d like more details or a hand with implementation, please don’t hesitate to reach out—I’m here to help! Sources: README.md | LiveKit's Series A: Infra for the AI computing era | README.md | README.md | Receiving and publishing tracks | LiveKit Docs