AI Creates Fake Obama

[PC2]

I imagine that this is just a more advanced version of how lip sync is done in some source games; it uses syllables and produces a lip shape with them. This'll be detecting the letters, just like a speech to text program works, and then filtering it through their shape engine which renders it onto a face.

I don't make games but i'm pretty sure faces and lips in games are heavily hand crafted and are changing the values of a graphical scaffold. Where as neural nets automatically train the pattern recognition and work directly at the pixel level.

[Sykol]

Thanks, AIbama.