Abstract

Phonation into tubes is used for voice training and therapy. In the present study, the formant frequencies were estimated from measurements of the acoustic pressure and the acoustic input impedance for a plexiglass model of the vocal tract (VT) prolonged by a glass tube. Similar transfer function measurements were performed with a human VT in vivo. The experimental results matched the mathematical modelling and confirmed the legitimacy of assuming rigid walls in mathematical simulations of the acoustic characteristics of an artificial VT model prolonged by a tube. However, this study also proved a considerable influence from soft tissues in the yielding walls of human VT cavities on the first formant frequency, F1. The measured F1 for the VT model corresponded to the computed value of 78Hz. The experiments in a human instead resulted in a much higher value of F1: about 200Hz. The results confirm that a VT model with yielding walls must be considered for mathematical modelling of the occluded or semi-occluded human vocal tract, e.g. prolonged by tubes or straws. This is explained by an acoustic-structural interaction of the vocal tract cavities with a mechanical low-frequency resonance of the soft tissue in the larynx.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call