Meta's Muse Generates Real-Time Talking Avatars From Any Portrait

Meta's Muse, an audio-driven Diffusion Transformer, streams 25fps avatar video from a single portrait with sub-second latency, outperforming Runway and HeyGen in head-to-head ratings.

Read the full story at AlphaSignal ↗

Timeline · 1 report

2026-11-05 18:36 · AlphaSignal
Meta's Muse Turns Any Portrait Into a Real-Time Talking Avatar

More stories

Meta adds real-time voice conversations and expressive avatars to Muse app — Meta Newsroom
Meta launches specialized AI models with Active Memory and Active Focus features — Techmeme
Meta introduces Spaces, a new way to collaborate with its Muse AI agents — TechCrunch AI
Meta adds 'spaces' to Muse AI for building custom app-like experiences — The Verge AI
Meta says current Muse model scores 47% on its Personal Superintelligence Index — Unite.AI
Meta distills Muse Wonder memory features into Muse and Meta AI, cutting forgetting 2.4x — AlphaSignal
Meta's Muse-trained models now power Meta AI for all users — The New Stack AI
Writer Tests Muse AI Agent for Birthday Trip Planning, Catches and Corrects Booking Error — CNET AI

Get the daily brief of stories like this at 6:30 every morning →