Search This Site

Perceptual, Acoustic, and Aerodynamic Characteristics of Front, Back, Dark, and Light Voice Qualities

Presenters Name: 
Melanie Turner
Co Presenters Name: 
Primary Research Mentor: 
Nicholas Barone
Secondary Research Mentor: 
12:45 - 1:00
Time of Presentation: 
2019 - 12:45pm to 1:00pm
Room 389
Presentation Type: 
Presentations Academic Category: 
Social Science
Grant Program Recipient: 
Harrison Undergraduate Research Grant

This study aimed to define and then compare two sets of vocal quality terms commonly implemented by singing professionals, front vs back and light vs dark, using acoustic, aerodynamic, and perceptual measurements, as well as a survey asking voice professionals to rate and define them. In phase one of the study, voice samples from members of University of Virginia singing groups were collected and analyzed using Phonatory Aerodynamic Systems (PAS) and Multidimensional Voice Protocol (MDVP). In phase two, voice professionals rated these recordings from front to back or light to dark to determine rater agreement for the terms and the degree of perceptual overlap between them. Listeners also qualitatively described how they define these terms. Phase 1 results showed that light voices had significantly different loudness, sub-glottal pressure, aerodynamic efficiency, relative average perturbation, jitter, and amplitude variation than front, back, and/or dark voices. There were no other significant differences in acoustic and aerodynamic properties between voice qualities. Results from Phase 2 showed that 75% inter-rater agreement was low for both voice quality pairings (44% for light/dark and 36% for front/back), with the highest agreement when rating samples from light to dark. These findings may be due to the lack of objective differences in the samples, the lack of experience of singers, or the inconsistent personal internal representations of these terms among singers and voice professionals which was indicated by the wide variety of anatomical, descriptive imagery, and other terms used to describe voice qualities in open-ended responses.