Sogou uses biometrics to develop AI news anchor for hearing impaired

Chinese search engine developer Sogou has launched an AI sign language virtual news anchor based on biometric technology, reported to be the first in the world, for China’s 27 million hearing-impaired people.

The AI anchor, known as Xiao Cong, is expected to vastly improve access to news and entertainment services, through integration of China’s Common Sign Language Dictionary, imitating hand gestures, facial expressions, lip movements, and mannerism of the sign language system with an 85 percent accuracy rate, the company says.

“We are thrilled to bring the AI sign language news anchor to hearing impaired viewers. And we hope to help them communicate better and get more engaged with society through our technological innovations,” says Wang Xiaochuan, Sogou’s CEO.

Last year, Sogou won first place in a facial landmark localization challenge at the 25th International Conference on Pattern Recognition (ICPR), a feature important for face alignment for biometric recognition. The company made headlines previously in 2018, scoring the highest true acceptance rate (TAR) in The MegaFace Challenge hosted by the University of Washington. In doing so, demonstrating its state-of-the-art R&D.

Xiao Cong is an upgrade from the company’s Vocational Avatar ‘Yanny’ first shown at the China Online Literature Conference in 2019, and is based on Sogou’s existing biometric technologies like 3D digital modelling, multi-modal recognition, facial recognition, animation and transfer learning.

Xiao Cong can be applied to news broadcasting, public services, special education and more. The company will partner with industry leaders, striving to make the technology accessible to all who need it.

Article Topics

AI  |  biometrics  |  expression recognition  |  facial recognition  |  gesture recognition  |  research and development  |  Sogou