Phonation threshold pressure and the properties of the vocal tract

Abstract

In a classic 1988 paper, Titze presented arguments based on the dynamics of the motion of the air through the glottis and its relation to the pressures there to describe how the presence of the vocal tract should affect the phonation threshold pressure. He argued that the action of the intraglottal pressures due to the vocal tract and the motion of the vocal folds would be in phase, and thus the presence of the vocal tract should lower the threshold pressure by an amount that depends upon its inertence. Since the inertence of the vocal tract depends directly upon its length and inversely upon its cross sectional area, these arguments set the stage for quantitative studies of the connection of the geometry of the vocal tract with threshold pressure in both mathematical and physical models. To this end, two sets of experiments were carried out in Erlangen with a physical model of the vocal folds and a vocal tract whose dimensions could be varied. One set of experiments focused on the relationship of the threshold pressure and its frequency with the cross sectional area (areas varied from about 2 cm2 to 12 cm2) and the other addressed the relationship of threshold pressure and its frequency with the length of the vocal tract (lengths varied in increments of 5 cm from about 4 cm to 54 cm). These measurements are compared with calculations done with the surface wave model and those done with a two-mass model.