Microsoft Research Asia released a new paper introducing VASA
Generate realistic videos based only on a single static image and audio track
Interesting article on Neowin 04/18/2024
Microsoft's new AI creates super-realistic talking-head deepfakes, and it made Mona Lisa rap
"Microsoft Research Asia released a new paper introducing VASA, a framework for generating lifelike talking faces. The researchers presented their model, dubbed VASA-1, that can generate realistic videos based only on a single static image and a speech audio clip."
Very impressive technology, but also very disturbing. It's getting to the point where we will not be able to differentiate between deep fake and reality.
Your thoughts?