Hajra Naz
Meta has expanded its artificial intelligence capabilities with a major consumer rollout. American users can now use the company’s proprietary assistant for real-time video calls. The feature relies on specialized avatar models designed to simulate live, face-to-face interaction.
Real-Time Avatar Architecture
The underlying engine relies on two core systems: Muse Realtime Avatar and Muse Realtime Voice. First previewed during Meta’s Connect conference, these combined models process speech instantly. They generate matching facial expressions and vocal inflections on the fly. Although the system doesn’t visually inspect the user through the camera, it listens for vocal cues and crafts synchronized responses.
Customization and User Engagement
Rather than relying on generic voice prompts, Meta pairs the experience with an interactive digital figure. The default character is a stylized mascot named Jolly. Users can easily reconfigure the appearance into different humans, animals, or fantasy avatars. Executives believe giving the assistant a physical form builds stronger user retention and deeper long-term engagement.
Read More: Meta Launches Its First-Ever Frontier AI Model: Muse Wonder
Public Demonstration and Product Roadmap
Meta’s Chief AI Officer, Alexandr Wang, highlighted the release by posting an interactive demo on X. In the clip, Wang conversed with a customized fluffy dog avatar. The digital assistant delivered quick, context-aware humor regarding corporate policy and social media posts.
This software update lays the foundation for upcoming physical hardware. Meta plans to introduce a dedicated wearable device called the Muse Charm before the end of the year. The compact hardware will host the animated avatar on a pocket-sized display, giving consumers a dedicated pocket companion.
