Google Adds Live Avatars to Gemini 3.8 Live for Enterprise Use
The feature synchronizes speech with lip movements and facial expressions, while custom avatar creation remains limited to allowlisted enterprise users.
Google introduced Live Avatar for Gemini 3.8 Live on Sept. 24, enabling enterprise users to hold conversations with animated personas whose mouth movements and expressions respond to generated speech.
The feature is generally available through Gemini Enterprise, while access conditions for individual customers are not specified. Organizations can use preset avatars, and custom avatars created from reference images are currently limited to enterprise users approved through an allowlist.
Live Avatar generates synchronized avatar video at 24 frames per second and produces audio at 24 kilohertz. It can switch among 97 languages while adapting the avatar's lip movements and facial expressions to match speech.
Gemini 3.8 Live accepts audio, live camera feeds and screen sharing simultaneously. It can also execute tools and application programming interface calls in the background while continuing the conversation.
The feature is aimed at customer service, interactive walkthroughs, kiosks and digital concierges. Generated audio and video include imperceptible SynthID watermarks.
“Live Avatar brings an interactive visual presence to conversational video agents across web, mobile, and interactive kiosks,” Fabien Blanc-paques, group product manager for Gemini Live, said.
“By pairing near real-time video generation with speech, the Live Avatar feature creates an experience that listens, sees, and speaks with a dynamic visual persona,” Shuo-yiin Chang, a research scientist, and CJ Zheng, a software engineer, said on behalf of Google's Gemini Audio Team.
The Gemini 3.8 Live model became available to developers through the Gemini API and Google AI Studio on Sept. 15, before the Live Avatar announcement.
