Cepstral Peak Prominence in Nondysphonic Children Using Praat and Analysis of Dysphonia in Speech and Voice

Purpose: This study examined the effect of age on cepstral peak prominence (CPP) in nondysphonic children between 3;0 and 17;11 (years;months) for two computer programs: Analysis of Dysphonia in Speech and Voice (ADSV) and Praat. Normative estimates for this population, the effect of sex, software, and stimuli on CPP, and the covarying impact of fundamental frequency (F0) were examined. Method: CPP and F0 were collected for 103 children (44 males, 59 females) from the vowel/a/and the all-voiced sentence “We were away a year ago,” within the following age ranges: 3;0–6;11, 7;0–10;11, 11;0–14;11, and 15;0–17;11. Effects of age, sex, stimuli, and software were examined using analyses of variance and post hoc means comparisons. The presence and strength of relationships between age, CPP, F0, and measures of CPP using ADSV versus Praat were evaluated using Pearson’s and Spearman’s correlations. Stepwise multiple regression analyses were computed to predict CPP from age and F0. Estimates of CPP normative cutoffs for Age × Sex groupings were also calculated. Results: Significant differences between 15; 0–17; 11 versus younger age chil-dren and a significant correlation between age and CPP were observed. Mean CPP values differed by sex, stimuli, and software. Age and F0 are significant predictors of CPP; however, the observed increase in CPP with increasing age in males is primarily due to the substantial decrease in F0 postpuberty. Signifi-cant effects of stimuli and software on CPP values were also observed.Conclusions: The findings support the hypotheses that CPP is correlated with age during the 3; 0–17; 11 span, with particular increases in postpubertal chil-dren. However, “normative” age-based expectations should be approached with caution since the general effect of age may be superseded by specific changes in F0 where a lowering of F0 is significantly associated with increases in CPP.