Gemini 3.8 Live now includes a Live Avatar feature that pairs near real‑time video generation with speech to provide a visual persona for native live dialogue models. The capability listens, sees, and speaks: it offers precise lip‑sync, natural facial expressions, and smooth turn‑taking to make virtual interactions more engaging.
Use cases
Live Avatar can be used to enrich customer service, deliver interactive walkthroughs, and otherwise enhance digital exchanges where visual presence improves comprehension and engagement. The feature supports asynchronous background tool calls, allowing the system to fetch data or trigger actions while maintaining an uninterrupted conversation.
Multimodal interaction and language support
By processing visual and audio inputs simultaneously, Live Avatar enables more natural multimodal conversations. It provides native multilingual speech‑to‑speech synchronization: the system dynamically adapts lip‑sync and expressions and can switch seamlessly across 97 languages without degrading video fidelity or causing visual drift.
Customization and brand identity
Organizations may choose from a library of preset avatars or create custom Live Avatars. From a high‑quality reference image, developers can generate a fully animated, responsive avatar that preserves reference likeness, brand styling, or character identity. Custom avatar creation is currently available only through enterprise allowlisting.
Asynchronous tool execution and continuity
Live Avatar is supported by Gemini’s advanced reasoning. Through asynchronous tool calling, the avatar can invoke tools and retrieve information in the background while continuing active dialogue, enabling it to handle complex tasks—such as checking in a hotel guest—without interrupting the conversational flow.
Trust and transparency
Live Avatar was developed with safeguards to respect identity and maintain transparency for AI‑generated content. All outputs produced by the AI are watermarked with SynthID; this imperceptible watermark is embedded in audio and video outputs to help detect AI‑generated content and reduce misinformation or misattribution.
Availability
Gemini 3.8 Live with Live Avatar is available in Gemini Enterprise. Developers can consult the API documentation to begin integration.
Summary
Live Avatar expands Gemini 3.8 Live’s conversational models with near real‑time visual presence, asynchronous background tool usage, broad multilingual support, and integrated safety measures such as SynthID watermarking, while custom avatar creation remains restricted to enterprise allowlisted customers.
