Neural Decoding of Chinese Sign Language with EEG Signals

As a basic language for communication in the deaf community, sign language contains rich body movement patterns, expression states and semantic information. The electroencephalogram (EEG) is a reliable physiological signal that can effectively decode motor information and expression states. The objective of this study was to investigate the neural characteristics of decoding Chinese sign language using electroencephalogram (EEG) signals. First, the study collected electroencephalogram (EEG) signals from 15 deaf college students while they watched sign language vocabulary videos and performed sign language actions. The collected EEG signals were then preprocessed, and the differential entropy and power spectral density features of the EEG signals were extracted. Finally, this paper used different classifiers to categorize six sign language EEG features. Experimental results show that the SVM-based classifier obtains the best classification results with an average classification accuracy of 66.27 %. This study shows that electroencephalogram (EEG) signals can effectively decode the neural features of sign language in the brain, providing some reference value for brain-computer interfaces (BCIs) to facilitate communication in the deaf community.