Google Unveils Gemini 3.8 Live with Live Avatar for Real-Time Visual Presence

The Live Avatar feature allows the AI to listen, see, and speak with precise lip-syncing and natural facial expressions. It supports native multilingual speech-to-speech synchronization across 97 languages, maintaining video fidelity and preventing visual drift during language transitions.

For enterprise use cases, the feature can facilitate interactive customer service or digital walkthroughs. The technology leverages asynchronous tool calling, which enables the avatar to trigger tool calls and fetch data in the background without interrupting the ongoing dialogue.

Organizations can choose from a library of preset avatars or create customized, fully animated avatars from a single high-quality reference image to maintain brand identity. While preset options are widely available, custom avatar creation is currently restricted to enterprises through an allowlisting process.