Spectral Aggregate of the High-Passed Fundamental Frequency and Its Relationship to the Primary Acoustic Features of Adductor Laryngeal Dystonia

Abstract

Objective: Currently, no clinically feasible objective measures exist that are specific to the signs of adductor laryngeal dystonia (LD), deterring effective diagnosis and treatment. This project sought to establish concurrent validity of a new automated acoustic outcome measure, designed to be specific to adductor laryngeal dystonia (AdLD): the spectral aggregate of the high-passed fundamental frequency contour (SAH f o ). Method: Twenty speakers with AdLD read voiced phoneme-loaded (more symptomatic) and voiceless phoneme-loaded (less symptomatic) sentences. LD discontinuities (defined as phonatory breaks, frequency shifts, and creak), the acoustic ramifications of laryngeal spasms, were manually identified. The frequency content of the f o contour was examined as a function of time, and content above 1000 Hz was summed to automatically calculate SAH f o . Multiple linear regression analysis was applied to SAH f o based on LD discontinuities and sentence type (voiced or voiceless phoneme-loaded). Results: The regression model accounted for 41.1% of the variance in SAH f o . Both the LD discontinuities and sentence type were statistically related to SAH f o . Conclusion: Results of this study provide evidence of concurrent validity. SAH f o is an automatic outcome measure specific to acoustic signs of AdLD that may be useful to track treatment progress.