1. Introduction

This chapter introduces the concept of cepstral peak prominence (CPP) and its value in measuring vocal quality, provides an overview of smoothed cepstral peak prominence (CPPs), and discusses factors that can affect CPPs measurements, such as computational algorithms and voicing detection. It also examines male-female differences in CPPs and presents previous studies that demonstrate the clinical application of CPPs for diagnosing voice disorders. Additionally, the chapter discusses Parkinson's disease and the specific motor speech disorder known as hypokinetic dysarthria associated with it.

2. Methodology

Participants with PD and healthy control participants were selected, with specific criteria for age, cognitive impairment, and voice disorders. Speech samples were collected and analyzed using a head-mounted microphone and a sound-attenuated room. Recorded sound files were transferred to Praat for CPPs measurement for both PD and healthy participants. A custom software plugin was utilized for CPPs analysis with specific default settings. A mixed effects ANOVA was computed with IBM SPSS, and CPPs cut-off values for the detection of dysphonia were examined.

3. Results

Table 2 presents the descriptive statistics of smooth cepstral peak prominence of vowel /a/ (CPPsvowel) and the Rainbow Passage reading (CPPsspeech) based on within-subject factors Task (Sustained Vowel, Rainbow Passage) and Gender (M, F) and between-subject factor of Group (PD, Control). The results of the mixed-design ANOVA revealed that there was a statistically significant simple main effect of Task, with producing sustained vowel presenting significantly higher CPPs than connected speech in both PD and control group. There were no significant interaction effects among Group, Task, and Gender, and no significant differences between males and females in the control group. Additionally, there were no interaction effects between Group and Task, between Group and Gender, and between Gender and Task, or significant simple main effect of Gender or differences between participants with PD and their healthy partners. When utilizing mean CPPs as the cut-off value for sustained vowel, 12 PD participants presented potential dysphonia concerns, while 17 presented concerns for connected speech.

4. Discussion and Conclusion

The key content includes an examination of smooth cepstral peak prominence in people with hypokinetic dysarthria associated with PD compared with healthy participants, as well as the variation of CPPs in speech tasks. Results showed that PD participants had lower CPPs in connected speech than in sustained vowels, and there was a significant difference in CPPs between males and females in PD participants. Also, the study suggested a consistent research protocol for comparable normative data and the use of multiple measures in addition to CPPs for the accuracy of clinical decisions. Smooth cepstral peak prominence is considered an accessible method for use in clinics, but more normative data are necessary for accurate cut-off values.