Association Between Vitamin D Status and Voice Features in the Thai National Health Examination Survey

Abstract

Objectives: Although vitamin D may affect human speech by influencing neuromuscular functions, related studies are limited and report discrepant findings because of small sample sizes and variations in voice parameters considered. This study investigated the relationship between vitamin D and voice parameters using the Extended Geneva Acoustic Parameter Set (eGeMAPs) in a randomly selected subpopulation (n = 480) of the Thai National Health Examination Survey. Methods: The /a/ voice samples provided by participants via smartphones were analyzed using the OpenSMILE Python package to extract 88 eGeMAPs features. Mass spectrometry was utilized to measure 25-OH-vitamin D (25(OH)D) levels. Subjects were divided into two groups. In Gr.1 (≤35 years, n = 98). Results: Stepwise multiple regression analysis identified 28 voice parameters significantly associated with 25(OH)D levels, independent of sex. The five most strongly associated voice parameters were mean spectral flux in voice segment, mean spectral flux, amplitude log ratio of the 3rd formant/F0, amplitude log ratio of the 2nd formant/F0, and equivalent sound levels. In Gr.2 (>35 years, n = 382), only the standard deviation of the falling slope of the fundamental frequency was associated with 25(OH)D levels, independent of sex. Conclusions: This work shows that vitamin D potentially influences neuromuscular functions associated with voice generation, particularly in younger individuals.