Two studies, both from researchers at the University of Oxford’s Department of Computer Science take different approaches in guiding robots to “read” speech and discern meaning without hearing a word. One team has developed a new artificial-intelligence system called LipNet that can be used to discern speech by lip-reading from silent video clips more effectively than professional lip-readers can. The study uses a data set known as GRID, which is made up of well-lit, face-forward clips of people reading three-second sentences. A second team has been working with Google DeepMind. using a series of 100,000 video clips taken from BBC television. These videos have a much broader range of language, with far more variation in lighting and head positions. Both experiments show AI vastly outperforming humans at lip-reading.