A lip-reading technology can now make fake videos of anyone. What’s more these videos will will look and sound like them. The creator of this technology, Supasorn Suwajanakorn, is currently a research resident at Google, though he created this as a part of his PhD thesis project at the University of Washington.
All the programme needs is a few photos and videos to learn from. Suwajanakorn demonstrated the tool onstage in Vancouver at the 2018 TED Conference. It uses a neural learning network to mimic mouth and teeth movements of a person from video footage. He has created fake videos of celebrities like former president Barack Obama using only images and videos that are readily available online.
Since all that the system needs is enough data to pick up on a speaker’s mannerisms by studying the subject’s teeth, lip movements, and jaw shape, it can be misused. It needs work before it can “fully model individual people” from head to toe.