24 July 2024

Lifelike talking faces


Explained at a Microsoft website:
We introduce VASA, a framework for generating lifelike talking faces of virtual characters with appealing visual affective skills (VAS), given a single static image and a speech audio clip. Our premiere model, VASA-1, is capable of not only producing lip movements that are exquisitely synchronized with the audio, but also capturing a large spectrum of facial nuances and natural head motions that contribute to the perception of authenticity and liveliness. The core innovations include a holistic facial dynamics and head movement generation model that works in a face latent space, and the development of such an expressive and disentangled face latent space using videos. Through extensive experiments including evaluation on a set of new metrics, we show that our method significantly outperforms previous methods along various dimensions comprehensively. Our method not only delivers high video quality with realistic facial and head dynamics but also supports the online generation of 512x512 videos at up to 40 FPS with negligible starting latency. It paves the way for real-time engagements with lifelike avatars that emulate human conversational behaviors.
Lots of video explanations at the site, followed by this disclaimer:
Our research focuses on generating visual affective skills for virtual AI avatars, aiming for positive applications. It is not intended to create content that is used to mislead or deceive. However, like other related content generation techniques, it could still potentially be misused for impersonating humans. We are opposed to any behavior to create misleading or harmful contents of real persons, and are interested in applying our technique for advancing forgery detection. Currently, the videos generated by this method still contain identifiable artifacts, and the numerical analysis shows that there's still a gap to achieve the authenticity of real videos.

3 comments:

  1. I'm getting old. I grew up with the first iteration of the interwebs. In the pre-Windows green and orange screen only days. What a revelation that was.

    But whatever they say, AI is only gonna lead to an enormous amount of shittification of the internet and society.

    Online help has already been replaced with AI help that is equally shitty as real help was. Pretty soon, phone operators will disappear and after that real people will disappear and will get stuck in a matrix of unhelpful AI interfaces that you simply can't get out of.

    Meanwhile, this beautiful internet with wonderfully curated sites like TYWKIWDBI will be engulfed by a tidal wave of AI generated bullshit.

    And the worst bit will be that important people like doctors will be value-engineered away by mistake-prone AI systems that no one will bear responsibility and liability for.

    When people say that AI is gonna kill us, it's because we will hand over our health care to systems we don't know how they work.

    /looks for cane to shake at clouds.

    ReplyDelete
  2. Just what the world needs more reason to distrust anything and everything we see and hear. But it's OK because we know somewhere, someone, is getting rich from it.
    xoxoxoBruce

    ReplyDelete
  3. Maybe if Microsoft was as concerned about fakery as they say they are they should have built in some sort of detection apparatus from the get-go. Not that hackers would eventually find a way around it, but make them work for it.
    Sandra

    ReplyDelete