Establishing Cepstral Peak Prominence (CPP) normative data for young adults in Ireland: the impact of gender, speech task and recording conditions

Abstract

Background: Cepstral Peak Prominence (CPP) is currently recommended as a robust measure of dysphonia in clinical voice assessment (Patel et al., 2018). Its use in clinic is limited due to its relative conceptual complexity and also the lack of normative data against which clinicians would compare disordered voice analysis data. Furthermore, CPP values can vary significantly with recording conditions, analysis software used, speech tasks, and the age and gender of the speaker. This study aims to establish normative data for CPP for young adults in Ireland and to determine the extent of CPP variability across genders, speech tasks and recording conditions. It aims to fill the gap with regard to CPP data collected remotely. Method: Thirty four participants aged between 18 and 24 years were recruited (28 female). Participants produced a number of speech tasks (sustained vowels and sentences with different segmental composition). Recordings were completed onsite in a sound insulated room using a head-mounted microphone, and remotely using the participants’ mobile phones. CPP values were extracted using Praat (Boersma & Weenink, 2023) and subsequently analysed using a complex mixed-effect ANOVA, with two within-subject factors (recording condition, speech task) and one between-subject factor (gender). Results: We report CPP data for young normophonic adults living in Ireland. Collection of data for the project is ongoing. Our preliminary results suggest a significant effect of speech task (lower CPP in connected speech compared to sustained vowels) while the effects of recording condition and gender were not found significant. Conclusions: Smartphones can be used to record audio data of adequate quality for acoustic voice analysis. Normative CPP data reported in the study could potentially contribute to the standardised use of CPP as a robust tool and facilitate its interpretation in the assessment and treatment of voice disorders, to evaluate treatment efficacy, and monitor progress.