Significance of unvoiced segments and fundamental frequency in infant cry analysis

Abstract

In this paper, significance of unvoiced segments and fundamental frequency (F0) in infant cry analysis is investigated. To find out the unvoiced segments from the infant cry F0 contour is used. For extraction of F0 contour, Teager Energy operator (TEO) based pitch extraction algorithm is used. TEO gives the running estimate of the signal energy in terms of its amplitude and instantaneous frequency. To quantify the importance of proposed features in infant cry analysis of variance (ANOVA) method is applied. It has been found that quantification of unvoiced segments and fundamental frequency in the cry, deliver information about the maturation of cry production system. In infant cry analysis, presence of high unvoicing ratio in a cry cannot be attributed to presence of pathology, like adult vocal fold pathological sounds.