Paper Presentation: Using Cepstral Peak Prominence Values in Research and the Voice Studio

Cepstral Peak Prominence (CCP) Analysis is used within clinical vocology for dysphonia research. Dysphonia is the medical terminology for a set of voice disorders characterized by hoarseness, breathiness, shakiness, etc. CPP has been demonstrated as a reliable metric for both sustained vowels and continuous speech. In the voice studio, CPP is a tool to help track the student’s vocal quality over time. Cepstral Peak Prominence and Smoothed Cepstral Peak Prominence (CPPS) are created from a cepstrum, also known as a “spectrum within a spectrum.” The cepstral peak and rahmonics—the cepstrum’s equivalents of fundamental frequency and harmonics, respectively—are used to determine a regression line. The larger the distance between the Cepstrum’s peaks and the regression line, the larger the CPP value is. Higher CPP values indicate a higher degree of periodicity in the voice, and subsequently a typical production of phonation. Understanding CPP values is imperative for anyone looking at current research about dysphonia. Researchers increasingly use this as an acoustical metric to certify perceptual values. However, there is a disconnect between the knowledge acquired within research and its practical application in the voice studio. Information surrounding cepstrum, calculating CPP values, and its mathematical concepts are often excluded from dialogs within vocal pedagogy. This presentation will focus on creating a firm understanding of these concepts, and explore how to perform these analyses with PRAAT through a guided demonstration. Vocal professionals will have an additional resource to monitor vocal health and information to pass to medical professionals if the need arises