A team of researchers in the USA have used Artificial Intelligence to identify and distinguish between baby cries. A convenient app for every parent who faces difficulties to make out a baby’s cry when it’s hungry, wet, in pain or simply just tired like the parent themselves trying to figure it out. The method is useful to stay at home parents, as well as in healthcare for doctors to recognize cries among ailing children.
The artificial method can differentiate between normal cry signals and abnormal cries of the baby, including the ones that signal on a possible underlying illness. The team used a cry language recognition algorithm which is based on automatic speech recognition that identifies various features of infant cries.
Although every baby cries differently, they share some common features as they mostly result share the same reasons. A major challenge faced during this study was to identify the hidden patterns in the cry signals of the babies, an obstacle for which the team has not been able to find an appropriate solution. “Like a special language, there are lots of health-related information in various cry sounds. The differences between sound signals actually carry the information. These differences are represented by different features of the cry signals. To recognize and leverage the information, we have to extract the features and then obtain the information in it,” reports Lichuan Liu, corresponding author and Associate Professor of Electrical Engineering and the Director of Digital Signal Processing Laboratory.
The team used compressed sensing which could allegedly process big data more effectively. The process of compressed sensing reconstructs a signal based on sparse data. This is especially beneficial in noisy environments, which is where a baby is expected to cry. Moreover, the algorithm can identify meanings of both normal and abnormal cry signals; the algorithm is independent of individual crier can be used in practical cases as well as to study various features and the urgency of baby cries. The researchers hence hope that the study can further find feet in medical care. “The ultimate goals are healthier babies and less pressure on parents and caregivers,” says Liu whose group has conducted the research study, “We are looking into collaborations with hospitals and medical research centers, to obtain more data and requirement scenario input, and hopefully we could have some products for clinical practice.”
The research has been published in the May issue of IEEE/CAA Journal of Automatica Sinica (JAS), which is a joint publication of the IEEE and the Chinese Association of Automation.