Muse can now talk back: live voice and expressive avatars roll out in the US

Meta rolled out real-time voice and expressive avatars in the Muse app on November 5, starting in the US on mobile, web and Mac. The avatar lip-syncs and reacts as Muse speaks, and users can interrupt mid-answer.

Reported byunite.ai (opens the original story)

In this brief: 4 sections 2 min read

- Tap the phone icon in the top-right of a chat to start a voice conversation
- Works on mobile, web and Mac; US users only at launch
- Powered by Muse Realtime Voice and Muse Realtime Avatar, previewed at Connect in September
- Three modes: voice-only, avatar beside the chat, or full-screen face-to-face
- Muse keeps working on tasks in the background while the conversation continues
- Hundreds of voice styles to choose from, including an energetic New Yorker and a pirate
- Avatar video is generated at 448x768, 25 fps, with about 870ms latency from end of turn to first response
- Generated video carries an invisible Meta Video Seal watermark
- Research demos show broader animation than what ships in the app today
- Meta says voice conversations are coming soon to AI Glasses and Muse Charm
- Muse Charm is expected to ship before the end of 2026
- No timeline given for voice outside the US

Read full analysis on unite.ai

Useful?