Acoustic and Aerodynamic Comparisons of Voice Qualities Produced After Voice Training

摘要

Characteristics of true vocal fold vibration such as the proportion of closed phase of vibration to open phase, longitudinal tension, and the amount of medial compression are used to define four conditions during Estill Voice Training. However, it is unknown whether trainees achieve these phonatory differences after training. Acoustic and aerodynamic measures were used to determine differences in Slack, Thick, Thin, and Stiff conditions. Twenty-four female speech-language pathology graduate students received training perceiving and producing these four conditions and volunteered to participate 3-5 months later. After a 20-minute refresher training, participants were recorded using the Phonatory Aerodynamic System with electroglottography and Computerized Speech Lab. Four Estill Voice Training experts independently categorized the voice quality productions. Aerodynamic and acoustic measures of productions classified by at least three of four experts as having the intended quality determined if measures differentiated among voice qualities and supported the hypothesized physiological concepts used in training at Bonferroni corrected P <= 0.0063. Results showed that Slack had low fundamental frequency (fo), low sound pressure level (SPL), and high vibratory instability; Thick had high subglottal pressure (Psg), high SPL, and high vibratory stability; Stiff had high airflow while Thin had lower Psg than Thick. Seven measures differentiated the four qualities with 88.1% accuracy while only Psg, airflow, and jitter were required to differentiate Thick, Stiff, and Thin with 88.7% accuracy. As acoustic and aerodynamic measures differentiated among voice qualities and supported the theoretical physiological characteristics used in training, they could be used to track accuracy during training.