Expressive visual animations now accompany Meta's real-time voice AI, permitting the system to interact with users as a virtual presence. This embodiment technology transforms the Muse Realtime Voice model into a visual agent that can engage in live video conversations rather than relying solely on text or audio.

The new functionality evolves Muse from its origins as a text-based chatbot into a multimodal assistant with a digital face. By integrating the Realtime Avatar, Meta aims to transition the interaction model toward more natural, face-to-face engagements using real-time animations synchronized with its voice AI.