Meta has started rolling out voice and video calling for its Muse AI agent in the United States, letting users hold real-time conversations with an animated avatar instead of typing. The company announced the launch on Thursday through a blog post and on X, where Meta's chief AI officer, Alexandr Wang, shared a demo of his own Muse in action.
The feature, called Muse Voice, was first shown at Meta's Connect developer conference in September and is now available to everyone in the US across the mobile app, web and Mac. Meta says support is coming soon to its AI glasses and to Muse Charm, a Tamagotchi-like wearable device it plans to ship before the end of the year.
Muse Voice features and availability
Muse Voice is powered by two new AI models Meta introduced with the feature: Muse Realtime Avatar and Muse Realtime Voice. Together they drive the digital character that talks with the user, animating in time with the words it says.
Unlike rival voice assistants, Meta's Muse leans on a customizable animated avatar, nicknamed Jolly, that users can reshape into a person, an animal or another character. Meta believes this anthropomorphic design could make people feel more connected to their AI agent, a bond that may deepen once live conversations become the norm.
The Verge's Jay Peters, who tested the feature on the iOS app, reported that his Muse agent responded quickly over voice, though it occasionally paused while looking things up, similar to a text chat. He also noted that the avatar animations can feel uncanny, especially after he set his agent to look like Mark Zuckerberg.
- Real-time voice and video conversations with the Muse AI agent
- An animated avatar that uses subtle expressions and gestures matching the conversation's tone
