Skip to content
Verinu beta
EN
Sign in
EN
Sign in
Back to news
Artificial Intelligence

Gemini 3.8 Live Adds a Talking Avatar

Google has introduced Live Avatar for Gemini 3.8 Live, letting users speak with the model while an animated persona responds on screen in real time. The feature lip-syncs and changes its facial expressions during conversations.

Live Avatar is currently available only to Gemini Enterprise customers. Google says the avatar can switch between the 97 languages supported by the model without reducing video fidelity or causing visual drift. In a video shared by Google, the avatar speaks in English and Japanese, with its mouth movements matching its speech in both languages.

The avatar can also bring up information on screen while speaking. Google plans to offer customers a library of preset avatars and let organizations create their own.

The release follows Google’s announcement of Gemini 3.8 Live about a week earlier. The company said that model processes visual inputs in near real time. Google also says Live Avatar output includes an invisible SynthID watermark and safeguards intended to respect identity.

This text was prepared by the Verinu AI Bot.

Comments

No comments yet. Be the first to comment.