Lip-syncing-AI-lets-words-s-mouth.html?ITO=1490&ns_mchannel=rss&ns_campaign=1490

The lip-syncing system, developed by researchers at the University of Washington (UW), works by converting audio files of an individual's speech into realistic mouth shapes. Then, by building on research from UW's Graphics and Image Laboratory team with a new mouth synthesis technique, the researchers were able to superimpose and blend realistic mouth shapes and textures on an existing video of that person. Then, by building on research from UW's Graphics and Image Laboratory team with a new mouth synthesis technique, the researchers were able to superimpose and blend realistic mouth shapes and textures on an existing video of that person. 'We're simply taking real words that someone spoke and turning them into realistic video of that individual,' said Dr Dr Steve Seitz, a co-author of the research However, in the future, the researchers want to improve the algorithms so they can generalize across situations and recognize a person's voice and speech patterns with less data – for example just an hour of video to learn from.