Review of the visual feature extraction research

Since the 1990s, we improved speech recognition results-based approach by lip-reading computer technology and integration of speech recognition technology. This method attracted the attention of some researchers, and made considerable progress, but there are still many problems to be a breakthrough. The article does a detailed introduction on the current status of lip-reading technology and describes some of the main relevant aspects of the traditional methods and new technologies. This focuses on the visual feature extraction and recognition technology, which aims to attract more researchers to be interested and concerned in this area of research. We hope that they can participate in the study of lip-reading technology to jointly promote the development of the field.