The scientists at the Oxford university have established an ultramodern artificial intelligence(AI) system that generates fake videos of an individual by using their motionless images and an audio clip.
This AI system also recognizes facial themes using face-recognition algorithms. As the audio clip plays, the AI system controls the mouth of an individual in the still picture so that it looks like the human being is speaking.
While the outcome is not totally perfect, the scientists trust that it can be done in real time and fake videos can be generated in a realistic manner.
“The application we’re thinking of is redubbing a video into another language,” said Joon Son Chung, one of the creators of the system, who is also a graduate student at the University of Oxford in the UK.
The audio from the news clips can be changed into a different language and the image movements can be updated accordingly. ”The new method could be useful for redubbing animated movies”, the new scientist reported.
It is not only the system that needs to be adjusted. The experts have already created fake videos, almost identical to the original ones.This new artificial intelligence system tool builds the action process in a simple and rapid manner in the edited footage with new audio.