Laryngeal perturbation analysis: minimum length of analysis window

Abstract

Laryngeal perturbation measures have been applied to the analysis of cycle-to-cycle changes in periodicity and amplitude of the acoustic voice signal for more than 25 years. Although such measures enjoy widespread clinical application, there is little agreement about basic methodology, including the length of signal to be analyzed. The purpose of this study was to examine changes in laryngeal perturbation measures as a function of length of signal analyzed in 18 subjects who complained of symptoms of possible laryngeal dysfunction. The results showed that as many as 190 cycles may be necessary before jitter asymptotes and as many as 130 cycles may be necessary before shimmer asymptotes. Pathological voices may require a longer analysis window for perturbation analysis than do nonpathological voices.