An update from the Wild Wild West of fake news technologies: A team of computer scientists have figured out how to make words come out of the mouth of former President Barack Obama — on video — by using artificial intelligence.
If you’ve been on the Internet at any point in the last year, there’s a good chance you’ve come across fake news articles. Soon we may see a wellspring of fake news videos.
As a team out of the University of Washington explains in a new paper titled “Synthesizing Obama: Learning Lip Sync from Audio,” they’ve made several fake videos of Obama.
Take for example, a time that he discussed the Pulse nightclub shooter and said “the investigation is ongoing, but we know that the killer was an angry and disturbed individual who took in extremist information and propaganda over the Internet.”
Obama did in fact say these words. But the computer scientists were able to make it look like he said them in places different from where he actually did — in this instance, in a different room at the White House. Audio and video images could be manipulated to make it look like he said them at a different time — say as a much younger Obama.
Multiple illustrations of the technology are in this video:
The researchers accomplished this feat not by cutting and pasting his body into different scenes — but by having a computer system called a neural network study hours and hours of video footage, to see how Obama’s mouth moves. Computers could then manipulate his lips and contort them to have any word come out.
In this experiment, the researchers did not make up things for him to say. But they did manipulate his mouth to seamlessly cut out entire sentences from speeches he’d given on camera, and even to change the order of sentences.
For example, the demo shows Obama talking about health care and saying: “Enrollment is open right now, but only until January 31st. Go to Healthcare.gov.”
In that speech, as Obama originally gave it, those two sentences were in reverse order — and there were many other words in between.
Researchers chose Obama because there’s so much high definition video available of him to study. In principle, the artificial intelligence that studied him could be used to make others lip sync too.