In the latest example of the initiative on artificial intelligence and deepfake, the researchers showed a new program that uses machine learning to allow the editing of videos to add, delete or change the words issued directly from the mouth of someone.
This work was carried out by scientists from Stanford University, the Max Planck Institute for Informatics, University of Princeton, and are, and shows the ability to edit what people say in videos and create a fake version realistic, and how easy it.
You can see a number of examples in the video clip below, so that the modified simple paragraph from the talk of people to say things you never say never in the video clip, while this program and it is in the research stage at the moment and is available as a treatment for consumers, but probably won’t be a long time until you reach the public!
To create hoax videos or modify video clips of the original, brings together scientists among a number of techniques, one voice, establishment of a three-dimensional model to match the paragraph that will be fabricated with the modern style of the person in the video clip, and in the tests was about 60% of the modifications look real.
Source: Text-based Editing of Talking-head Video