The animated persona lip-syncs across 97 languages and will ship with preset and custom avatar options.
Google launched Gemini 3.8 Live with a feature called Live Avatar, which lets users hold conversations with the model while an animated persona responds in real time. The avatar lip-syncs and displays facial expressions, and Google currently limits access to Gemini Enterprise customers.
Google unveiled the Gemini 3.8 Live model about a week before the avatar launch, describing it as processing visual inputs in near real time. Google says the avatar moves between 97 languages without degrading video fidelity or introducing visual drift.
Enterprise teams gain a conversational interface that pairs real-time speech with on-screen information the avatar can pull up while talking. Building a custom avatar gives organizations a way to brand agent interactions without assembling separate video pipelines for each supported language.
Google plans to offer a library of preset avatars and to let organizations create their own. Output from Live Avatar carries the invisible SynthID watermark along with safeguards that Google says respect identity, so buyers should watch for availability beyond enterprise accounts.
What matters
- Google’s Live Avatar responds in real time with lip-sync and facial expressions for enterprise customers.
- Enterprise teams can deploy avatars that switch among 97 languages without visible video drift.
- Google will let organizations build custom avatars and stamps output with SynthID watermarks.
Why it matters
Google will let organizations build custom avatars and stamps output with SynthID watermarks.
This GenAI News article was prepared in original wording using reporting and materials published by The Verge. Source reference: https://www.theverge.com/tech/1000328/google-gemini-ai-live-avatar-face.
Drafted by the GenAI News review pipeline.