Application of Incremental SVM Learning for Infant Cries Recognition

Crying is the infant´s first communication. Before learning how to express the emotions or physiological/ psychological requirements with language, infants usually express how they feel to parents through crying. Infants have very strong curiosity and learning ability to mimic parent´s habits and reactions, thus, infants will express their needs by changing crying. To accurately determine the meanings of the newborn cries, we proposed a novel method that constructs personalize crying models for different infants. In this paper, the incremental learning support vector machine (SVM) is used for infant cry recognition system. Fifteen features were extracted from crying. Four significant features were finally selected to identify three kinds of crying. Experimental results show that the proposed method achieves higher accuracy than that of our prior work.