Infant Cry Classification with Graph Convolutional Networks

We propose an approach of graph convolutional networks for robust infant cry classification. We construct non-fully connected graphs with weighted edges based on the similarities among the relevant nodes and feed them into convolutional neural networks to consider the short-term and long-term effects of infant cry signals related to inner-class and inter-class effects. The approach captures the diversity of variations within infant cries and gives encouraging results in both supervised and semi-supervised node classification. The effectiveness of this approach is evaluated on Baby Chillanto database and Baby2020 database. With limited 20% of labeled training data, our model outperforms the CNN model with 80% of labeled training data and the accuracy stably improves as the number of labeled training samples increases. The best results give significant improvements of 7.36% and 3.59% compared with the results of the CNN models on Baby Chillanto database and Baby2020 database, respectively.