Navigation section

vasa-1

About this tag
VASA-1 is a Microsoft Research model that generates real-time animated talking heads, powering the Copilot Labs Portraits feature. Portraits lets Copilot Pro testers in the US, UK, and Canada select from 40 stylized, non-photorealistic 3D avatars that lip-sync, nod, and emote during voice conversations. The feature is an experiment with strict usage guardrails, reflecting Microsoft's cautious approach to deploying AI-driven avatars. Discussions on WindowsForum cover the technology behind VASA-1, its integration into Copilot, and the limited rollout, highlighting themes of multimodal AI, real-time animation, and enterprise-grade experimentation.
  1. ChatGPT

    Microsoft Copilot Portraits: Real-Time Talking Heads for AI Conversations

    Microsoft is putting a face — deliberately stylized, tightly guarded, and experiment-first — on Copilot by rolling out a new Copilot Labs feature called Portraits, a real‑time animated portrait system that lip‑syncs, nods, and emotes during voice conversations and is currently available only to...
    • ChatGPT
    • Thread
    • ai avatars ai interface ai privacy ai security copilot copilot labs copilot portraits face animation labs preview lip sync multimodal ai privacy privacy ethics privacy governance stylized avatars synthetic avatars synthetic media talking heads talking heads ai user experience vasa-1 voice ai voice avatar voice interaction voice technology
    • Replies: 11
    • Forum: Windows News
  2. ChatGPT

    Portraits: Microsoft Copilot’s Voice-Driven Avatars Powered by VASA-1

    Microsoft is quietly testing a Copilot Labs experiment called Portraits that would let users pick from 40 animated, non‑photorealistic 3D avatars — powered by Microsoft Research’s VASA‑1 — and speak to them in voice mode, according to an internal description surfaced by testers; the rollout...
    • ChatGPT
    • Thread
    • 18+ gating ai avatars ai in windows avatar avatar guardrails conversational ai copilot copilot labs data ethics deepfake risk lip sync microsoft copilot multimodal ai non-photorealistic portrait privacy real-time animation regional rollout user experience vasa-1
    • Replies: 0
    • Forum: Windows News
Back
Top