On November 6, Zuckerberg announced on Threads that the personal AI assistant Muse had officially launched real-time voice conversations and a virtual avatar. The feature is first rolling out to U.S. users across mobile, web, and the Mac app. Users can tap the phone icon in the top-right corner of the app and talk to it as if making a phone call, interrupting at any time while it is speaking. This means Muse is no longer just text in a typing box, but a conversation partner that can respond and talk back.
One model suite handles both mouth and face It is powered by two models: Muse Realtime Voice and Muse Realtime Avatar. The former generates both "what to say" and "how to say it," while the latter uses the same data stream to add facial expressions, lip sync, and movements in real time, keeping sound and visuals in sync. The system can stream portrait video at 448×768 and 25 frames per second, with about 870 milliseconds from when the user finishes speaking to the first audio-visual response. For interaction, users can use voice only, keep the avatar beside the chat, or switch to full-screen "face-to-face" mode. The default cartoon character Jolly can be replaced with people or animals, and there are hundreds of voice styles, from "New Yorker energy" to "pirate speak."
Aimed at "hands-busy" scenarios Meta positions it as a more natural personal agent: when walking or when your hands are busy, you can assign tasks by voice, and Muse keeps working in the background. It can also be used for brainstorming, travel planning, or practicing for interviews and presentations. The company also demonstrated making the avatar dance, then sharing it to Instagram, Facebook, and WhatsApp with one tap. When AI grows a face and a voice, the companion-chat business will probably be pushed to compete over "human-likeness."
