This video blows me away:
Basically someone used a Generative Adversarial Network (GAN) software package to do a deep fake face swap, as well as voice swap, of The Donald (and others) for this music video.
The effect, and the audio, isn't perfect, but speaking as someone who spent a fair amount of time in the video production world, I can tell you that doing this kind of face swap, even five years ago, would have taken a small, dedicated team hundreds of hours to pull off.



Reply With Quote


