Effects of voice training in normal-speaking women: Videostroboscopic, perceptual, and acoustic characteristics

Abstract

Several aspects of normal voices were investigated in eight female speakers (20-31 years) before and after voice training. The subjects were recorded on audio-tape during running speech and sustained-vowels and examined with video-fiberstroboscopy. Glottal closure pattern was evaluated by voice clinicians who analysed video-sequences using a 6-point rating scale. Audio-perceptual analyses of running speech and vowels were performed by experienced listeners who rated each of seven deviant voice quality parameters on a continuous scale. Acoustic analyses of fundamental frequency and of the relative level of the fundamental vs. the first formant were performed. Among the results, the posterior glottal chinks decreased somewhat in size after voice training, but were still observed in most subjects. For running speech, the degree of perceived breathiness, hypofunction, and vocal fry decreased while sonority increased significantly. Mean sound pressure level was raised by about 2 dB. and the mean fundamental frequency decreased significantly after training. For sustained vowels, the level of the first formant increased as compared to that of the fundamental. The results support the hypothesis that posterior glottal chinks found in women with trained voices are primarily caused by anatomical structures of the female larynx and not by vocal behavior.