OH, THE INTERNET. Why must you mess with our minds so much? Why has someone decided to make a Max Headroom-like Obama speechmaking clone head that looks and sounds a hell of a lot like the real thing?
Because of research, of course. This is about neural networks and machine learning and is the work of the University of Washington. A post on the University's news site says that Obama is a lip-syncing model of the physical ex-president and that it can be applied to any audio and give a pretty good damn reading of it.
The tool could be used in a very exploitative manner if it fell into the wrong hands and onto the wrong face, and there is something more than a little creepy about it.
It's groundbreaking stuff though, unless you count what Cassetteboy does, and the puppet masters are proud of what they have done. They should be, as the system can trawl through hours of tedious video - so you don't have to - to learn the nuances of facial expressions. And is happy to be doing it.
"These type of results have never been shown before," said Ira Kemelmacher-Shlizerman, an assistant professor at the UW's Paul G. Allen School of Computer Science & Engineering.
"Realistic audio-to-video conversion has practical applications like improving video conferencing for meetings, as well as futuristic ones such as being able to hold a conversation with a historical figure in virtual reality by creating visuals just from audio. This is the kind of breakthrough that will help enable those next steps."
A paper on the puppet ex-president is to be presented at SIGGRAPH (special interest group on Computer GRAPHics and Interactive Techniques) 2017 in early August.
A video is on YouTube you can bask in fake Obama there, maybe try and recognise him so that you can be prepared for the day when hackers break into the internet with a fake announcement from him. µ
Despite local protests
Remember crime does not pay people
You're crap but we couldn't have done it without you