Extracting dynamic parameters from speech movement data

Velocity differences in laryngeal adduction and abduction gestures

The periodic repetitions of laryngeal adduction and abduction gestures were uttered by 16 subjects. The movement of the cuneiform tubercles was tracked over time in the laryngoscopic recordings of these utterances. The adduction velocity and abduction velocity were determined objectively by means of a piecewise linear model fitted to the cuneiform tubercle trajectories. The abduction was found to be significantly faster than the adduction. This was interpreted in terms of the biomechanics and active control by the nervous system. The biomechanical properties could be responsible for a velocity of abduction that is up to 51% higher compared to the velocity of adduction. Additionally, the adduction velocity may be actively limited to prevent an overshoot of the intended adduction degree when the vocal folds are approximated to initiate phonation.

Identifying a differential equation for lip motion

A second order homogeneous differential equation is fitted to lip trajectories during speech at fast, normal, and slow speaking rates. The objective is to characterize lip motion in terms of biomechanical parameters such as stiffness, damping, and applied external forces. The results show that, at fast speaking rates, the lip behaves like a simple mass-spring oscillator moving at its natural frequency, and with low variability across repetitions. At slow speaking rates, on the other hand, the lip is under a stronger external control to conform to the required trajectory, which introduces a larger variability across repetitions. The case of normal rate falls between the fast and slow rate cases.

Intentional changes in sound pressure level and rate: their impact on measures of respiration, phonation, and articulation

The purpose of the study was to compare the effects of changing sound pressure level (SPL) and rate on respiratory, phonatory, and articulatory behavior during sentence production. Ten subjects, 5 men and 5 women, repeated the sentence, "I sell a sapapple again," under 5 SPL and 5 rate conditions. From a multi-channel recording, measures were made of lung volume (LV), SPL, fundamental frequency (F0), semitone standard deviation (STSD), and upper and lower lip displacements and peak velocities. Loud speech led to increases in LV initiation, LV termination, F0, STSD, and articulatory displacements and peak velocities for both lips. Token-to-token variability in these articulatory measures generally decreased as SPL increased, whereas rate increases were associated with increased lip movement variability. LV excursion decreased as rate increased. F0 for the men and STSD for both genders increased with rate. Lower lip displacements became smaller for faster speech. The interspeaker differences in velocity change as a function of rate contrasted with the more consistent velocity performance across speakers for changes in SPL. Because SPL and rate change are targeted in therapy for dysarthria, the present data suggest directions for future research with disordered speakers.