A SIGN LANGUAGE IDENTIFICATION SYSTEM TO SUPPORT DISABILITY LANGUAGE USING SUPPORT VECTOR MACHINE

Abstract

Communication between disability language individuals and non-disabled individuals often faces significant challenges. Currently, there are many people with disability language both worldwide and in Vietnam, and this calls for a useful solution to help individuals with speech impairments communicate more easily. This paper proposes a system to help disability language individuals communicate more easily with non-disabled people. The proposed system included a sensor glove to measure finger movement signals. The measured signals were preprocessed to remove noise and artifacts. The processed signals were then used to identify the letters corresponding to the gestures. Finally, the sounds corresponding to the identification letters were played through a speaker. The system achieves a 99.67% accuracy rate in sign language identification. These promising results suggest that the system could be applied in real-world scenarios to assist individuals with speech disabilities.