Google officially announced Gemini 3.8 Live Avatar Inside Gemini Enterprises. Novel introduction Interactive avatars with instant voice responsesproviding a more immersive experience and more natural, authentic conversations. The goal is to provide digital interactions that are more accessible, natural, and useful. Let’s take a closer look at how avatars work.
Interactive avatars with faces, voices and expressions
As we expected, Live Avatar introduces personalized avatars combined with near-real-time video generation and voice, providing Listening, Seeing and Communicating Through a dynamic character. Precise lip sync helps deliver more realistic and convincing feedback, and natural expression and smooth turn-taking. As you can see from the video, responses require a short wait, and the AI is able to visualize its surroundings and use advanced reasoning skills to have a more natural conversation.
While a conversation is active, Gemini can use the necessary tools and retrieve material in the background. Simply put, it can Manage complex tasks Maintain a smooth, uninterrupted flow of conversations. Obviously, there’s no need to reveal yourself in the video: as you can see below, you can interact with Gemini using just your voice.
Different options and features
Live avatars are also available Native multi-language voice synchronizationdynamically adjust lip sync by switching between 97 available languages. See how it works here:
Also, it’s okay Customize your avatar Also use reference images. We remind you that Gemini 3.8 Live with Live Avatar is available in Gemini Enterprise.
Android Drop launches five new useful features in September 2026: Motion Assist, Gemini, and more
What do you think? Are avatars with similar functionality actually useful in certain use cases? As always, let us know in the comments below.