Sign language is a silent language that expresses information through gestures, facial expressions, and body movements. For the deaf and hard of hearing, sign language is their main way of communication. However, due to the particularity of sign language, it is very difficult for computers to understand and recognize sign language. Therefore, it is of great practical significance to study how to use artificial intelligence technology to achieve sign language recognition. This chapter proposes a CNN-ABC method based on optimized combination blocks to recognize Chinese sign language, employing batch normalization, pooling, Leaky ReLU, and Dropout techniques. Experimental results show that our CNN-ABC method achieves an overall accuracy of 93.32 ± 1.42%, which is superior to 15 state-of-the-art methods compared.