Google DeepMind published Gemini 3.8 Live with Live Avatar on September 24, pairing near real-time video generation with speech to produce animated AI personas with lip-syncing.
The feature supports asynchronous tool calling for background data fetching during conversations, native speech-to-speech across 97 languages, and simultaneous vision and audio input. Organizations can build custom avatars from a reference image to preserve brand styling, subject to enterprise allowlisting, and all output is watermarked with SynthID to keep AI-generated content detectable.