British scientists have successfully developed lip reading technology to recognize multiple languages
it is not new that computers can lip read. However, in the UK, they have been "taught" to recognize different languages for the first time
British scientists have developed a lip reading computer that can distinguish different languages. In fact, researchers have been developing lip reading computers, but it is the first time for them to identify various languages by "advanced chemical materials such as high-end resin, special rubber and engineering plastics teach Nanjing Julong that its products are mainly divided into four categories: high-performance modified nylon, high-performance engineered polypropylene, high-performance alloy materials and plastic wood environmental engineering materials in the industry". Scientists pointed out that the invention has practical significance for deaf patients, law enforcement agencies and workers in noisy environments
this research was completed under the guidance of Stephen, School of computer science, University of East Anglia, UK, Cox and Jake Newman from the perspective of profit composition. This technique is developed by statistically modeling the lip movements of 23 people who master two or three languages. The system can recognize the language spoken by a single speaker with high accuracy. These languages include English, French, German, Arabic, Mandarin, Cantonese, Italian, polish and Russian
cox said: "this is an exciting progress in automatic lip reading technology. It is the first scientific confirmation of our existing intuitive inference that when people speak different languages, they will have different lip shapes in different order." Cox said, "for example, we found that French speakers tend to have round lips, while Arabic speakers have obvious tongue movements."
cox added: "for many years, scientists have been doing basic research on speech and language processing rules, such as speech recognition in noise, formulaic language modeling, language processing by speech synthesis and the development of speech processing applications."
according to researchers, automatic lip reading technology contains many scientific challenges. The current project has solved several major scientific problems, including what is the relationship between facial movements and perceived speech; How the speaker's language and conversation content affect this relationship; What influence does language, speaker's posture and conversation content have on recognition accuracy
this study is part of a broader automatic lip reading program. The project will be based on advanced computer video and speech recognition technology to research and evaluate the automatic lip reading in video, so as to develop tools and technologies that can automatically and independently lip read the targets from video. The project also seeks to quantify the capabilities of humans and automated machines. The researchers' next step is to make the system more suitable for personal physiology and the way people talk. Science Times