Meta has rolled out Muse Voice to everyone in the United States, letting users call and video chat with Muse, its consumer AI agent, through an animated avatar. The feature, first previewed at the Connect developer conference in September, runs on new Muse Realtime Avatar and Muse Realtime Voice models. It works on mobile, web and Mac, and Meta says it will come later to AI Glasses and the upcoming Muse Charm wearable.
A face for the assistant
The avatar is the point. While rivals offer voice chat, Muse appears as an animated character that speaks with subtle expressions and gestures matched to the tone of the conversation. The default character is nicknamed Jolly, and users can customise it into something that looks more like a person, an animal or another character entirely.
Meta announced the rollout in a blog post, and chief AI officer Alexandr Wang shared a demo on X. In the video, Wang jokes with his Muse about his habit of posting memes, and his avatar, a fluffy white dog, suggests adding disclaimers to memes that could be mistaken for Meta endorsing something risky.
The demo itself was part of the pitch. Wang’s Muse did not just answer questions; it joked back, matching his tone and keeping the conversation going. That is the experience Meta wants to sell: not a voice that reads answers, but a character that holds a conversation.
Customisation is part of the appeal. Users who want their assistant to look like them can shape the avatar that way, while those who prefer something playful can pick an animal or another character. The choice is cosmetic, but it is also the point: the assistant becomes something personal.
The hardware plan behind it
Muse Voice arrives just before Meta’s planned launch of a Tamagotchi-like wearable for Muse before the end of the year. When that device arrives, the digital avatar will appear on its small screen, letting users talk to the agent and have it reply or carry out tasks.
The video chat rollout is the on-ramp: get people talking to an avatar now, then put that avatar on their wrist. Meta has bet on wearables before, and the Muse Charm is the company’s clearest attempt yet to make its AI agent something people carry with them.
The strategy is familiar. Meta has spent years trying to make its AI feel like a companion rather than a tool, and an avatar that reacts with expressions is a step further than a voice alone. Whether users want a face on their assistant is the open question, and Meta is betting that the answer is yes.
Our read
This is Meta playing to its strengths. No other major assistant puts a customisable animated avatar in front of the user, and the Jolly default is designed to be charming rather than uncanny. The risk is that charm wears thin: an avatar that gestures and emotes can feel engaging in a demo and distracting in daily use.
The timing is also defensive. Muse has been rolling out fast, it recently launched in Singapore with local features, and Meta needs the assistant to feel distinct from ChatGPT, Gemini and Claude. A video-calling avatar is a real difference, and it gives Meta a reason for users to open Muse first.
Why It Matters
Voice assistants have been faceless for a decade, and Meta is testing whether giving one a face changes the relationship. If users bond with an avatar, Muse becomes a daily companion rather than a search box, and that is exactly the habit Meta wants. If they do not, the avatar is a novelty that fades once the demo ends.
Background
Muse is Meta’s consumer AI agent, and the company has been expanding it through 2026, including a launch in Japan and new coding-agent capabilities after buying Conductor. Muse Voice was first shown at Connect in September, alongside the company’s wider push into AI glasses and wearables.
Source: TechCrunch’s report
0 Comments