Muse now talks back in real time, with an avatar that reacts while you speak

Muse, Meta's personal AI agent, now holds real-time voice conversations. It's rolling out in the Muse app on phones, the web and the Mac. Tap the phone icon, top right, and start talking.

Slide text

Tap the phone icon, top right, in the Muse app on mobile, web and Mac.

Meta Newsroom · “Give Your Muse a Voice” · Nov 5, 2026

You no longer need the exact prompt. Start a trip plan with a city you've always wanted to see, and Muse picks up the thread and starts working while you keep talking, even out on a walk.

Voice turns Muse into a rehearsal partner. Ask it to play the hiring manager for your job interview, or run through a presentation or a pitch. Answering out loud, in the moment, feels closer to the real thing.

Slide text

Muse plays the hiring manager so you can rehearse

Or run through a presentation or a pitch.

Meta Newsroom · “Give Your Muse a Voice” · Nov 5, 2026

Muse's avatar now matches the conversation's tone with small expressions and gestures, and reacts while you talk. You pick how much of it you see: voice only, beside the chat, or full screen.

You also choose how your Muse sounds, from hundreds of voices. Want the energy of a New Yorker, or the flair of a pirate? Pick a style for your mood, and change it whenever you like.

Replies start in under a second after you stop speaking, with the avatar's voice and video arriving together. And you can interrupt Muse mid-sentence, the way you would a friend, and it keeps up.

In side-by-side live calls, people preferred Muse's avatar to Runway Characters and HeyGen LiveAvatar, overall and on every measure. Only on mannerisms, against Runway, did the two come out even.

Slide text

People picked Muse’s avatar over Runway and HeyGen

Mannerisms against Runway Characters were a statistical tie.

Meta AI Research · “Bringing Your Muse to Life” · Sept 23, 2026

Every frame of video the avatar generates carries a hidden watermark, from Meta Video Seal, so generated footage stays traceable. It adds no delay to the conversation.

Voice and avatars are rolling out in the US first, and voice is coming soon to AI Glasses and Muse Charm. If you try one thing, rehearse an interview out loud. That's where talking beats typing.