| Abstract | Objective: To verify whether there are differences in cepstral peak prominence (CPP) and smoothed cepstral peak prominence (CPPS), obtained in different speech tasks, as a function of the presence and severity of voice disorder (GG- general grade); as well as to analyze whether there is a correlation between CPP, CPPS and GG in different speech tasks.
Method: We analyzed 258 vocal samples of the vowel /a/ and number counting, including 160 dysphonic and 98 non-dysphonic voices. The counting samples were edited in three different durations: counting from 1 to 10, from 1 to 11 and from 1 to 20. Five speech therapists performed the auditory-perceptual judgment (APJ) of the GG. We used an automatic extraction script from the Praat software to extract the CPP and CPPS from all the vocal samples.
Results: There was a difference in CPP and CPPS between dysphonic and non-dysphonic individuals, regardless of the speech task used, with lower values in the dysphonic group. There was a difference between CPPVowel vs. CPP10, CPP11, and CPP20 values in the general group and non-dysphonic individuals, while CPPS showed a difference between all speech tasks in all groups investigated. Only the CPPS showed differences between all the speech tasks depending on the degree of vocal deviation. There was a moderate negative correlation between CPPVowel, CPP10, CPP11, and CPP20, and GG, as well as a strong negative correlation between CPPSVowel, CPPS10, CPPS11, and CPPS20 and GG.
Conclusion: There are differences in the CPP and CPPS obtained in different speech tasks, depending on the presence and severity of voice disorder. CPP and CPPS values are different between dysphonic and non-dysphonic individuals in all speech tasks. There is a moderate negative correlation between the CCP in the different speech tasks and the GG, and a strong negative correlation between the CPPS in the different speech tasks and the GG.
|
|---|