Gemini Live Avatar adds a lip-sync face for enterprise customers
Google's Gemini 3.8 Live update adds Live Avatar, an animated face that lip-syncs during a conversation. It is open only to Gemini Enterprise customers. Customers can pick a preset face or build their own. Google says the output carries an invisible SynthID watermark and that the face can switch among 97 languages without losing video fidelity.
Artificial Intelligence··Night
An enterprise face, not a public one
Google's Gemini 3.8 Live update adds Live Avatar, an animated face that lip-syncs during a conversation. The Verge says it is available only to Gemini Enterprise customers. The Register describes the same feature as an avatar that can present animated characters and realistic human simulacra to lip-sync Gemini's machine-generated speech, and it quotes Google DeepMind research scientist Shuo-yiin Chang and Gemini software engineer CJ Zheng calling Live Avatar a capability for enterprise agents that process visual and audio inputs together.[1], [2]
Presets, a custom face, and a watermark
Customers can pick a preset face or build their own, and Google says the output carries an invisible SynthID watermark. The face follows the Gemini 3.8 Live model Google described about a week earlier.[1]
97 languages, as Google's fidelity claim
Google says the avatar can switch among the 97 languages it supports without degrading video fidelity or introducing visual drift. That is the company's claim, not an independent measurement. The Register says the face can be a cartoon-like animated character or a lifelike human likeness, lip-syncing speech the model generates, and that it is offered for enterprise agents rather than as a consumer app. The Register's quotation comes from a Google blog post the two engineers signed.[1], [2]
