Human vocalizations contain a number of physiological and emotional clues and are viewed today as potential digital biomarkers. In particular, infant crying plays a central role, as it is the newborn’s primary means of communication and can provide information about health status, physiological needs, and emotional conditions. In this context, the scientific literature has proposed various methods for extracting acoustic features from infant vocal signals. In recent years, the evolution of artificial intel-ligence techniques, and machine learning in particular, has enabled the development of increasingly advanced approaches for the automatic analysis of crying, with promising ap-plications in clinical and care settings.