The new breakthrough by a team of American researchers can now solve the infant cry problem. The researchers devised an AI technique that has the potential to recognize and differentiate the pattern in a normal and an abnormal baby cry. For example, the technique can find out illness through the crying pattern of the baby.
The technique which is developed on a cry language recognition algorithm is capable of helping parents at homes as well in healthcare centers; assisting doctors to understand the crying pattern of normal and sick children.
During the research, the team stuck to compressed sensing in order to study and classify the signals of a crying baby. The process assisted the researchers to process bigger data more efficiently. After that, the researchers created a fresh cry language recognition algorithm which could differentiate between a normal and abnormal baby cry in a noisy environment.
The algorithm is different for every baby, as it can be used in a broader sense in real-life situations, as a method to detect and classify different cry features and the main cause behind the crying of a baby.
The team is very positive about their research work. The findings of the study have the capability to be used in various other medical care circumstances in which decision making plays a vital role. The research work was showcased at IEEE/CAA Journal of Automatica Sinica (JAS).