Google Launches Gemini 3.8 Live with Live Avatar for Enterprises

On September 24, 2026, Google launched Gemini 3.8 Live with Live Avatar, designed to enhance conversational AI experiences for enterprises. The Gemini Audio Team claims that the Live Avatar feature enables a more natural and intuitive interaction between enterprise agents and users. This product supports multilingual speech-to-speech synchronization in 97 languages, broadening its applicability across diverse markets.

Custom avatar creation is currently restricted to enterprise allowlisting, ensuring that only authorized users can utilize this feature. Additionally, all output generated by Gemini's AI products is watermarked with SynthID, providing a layer of content authenticity and traceability. This launch follows the recent introduction of Gemini 3.8, which included significant advancements in text-to-speech capabilities. The integration of Live Avatar into Gemini's offerings marks a notable shift towards more interactive and personalized AI solutions for enterprise applications.