In a significant advancement for AI interaction, Google has launched the Gemini 3.8 Live update, enabling users to engage in real-time conversations with an animated AI avatar. This feature, dubbed the “Live Avatar,” not only lip-syncs but also showcases various facial expressions during interactions. Currently, it is accessible exclusively to customers of Gemini Enterprise.
According to Google, the Live Avatar can seamlessly transition between 97 supported languages without compromising video fidelity or introducing visual inconsistencies. A recently shared demonstration illustrates the avatar conversing fluently in both English and Japanese, with mouth movements perfectly synchronized to its speech in each language. Additionally, it can retrieve information on-screen while communicating.
This unveiling comes shortly after Google introduced its innovative Gemini 3.8 Live model, which is designed to process visual inputs almost instantaneously. While users will have access to a library of pre-designed avatars, organizations will also have the option to create custom avatars. The output from the Live Avatar will feature an invisible SynthID watermark, ensuring proper identity representation.
Editor’s Take
The introduction of the Live Avatar marks a crucial milestone in AI communication, offering enriched user experiences and innovative possibilities for businesses. By enabling real-time, multilingual interactions, this technology not only enhances customer engagement but also sets a new standard for virtual communication tools. As developers harness these capabilities, the potential for customized solutions could reshape various industries and improve user accessibility to AI-driven systems.
Source: www.theverge.com
