Abstract
Objectives
To investigate whether there are age differences in acoustic voice quality, Mandarin lexical-tone production, and within-speaker tone stability in Mandarin-speaking healthy women.Methods
A total of 220 women aged 18-84 years participated in a cross-sectional study and completed sustained vowel, maximum phonation time, connected speech, and Mandarin lexical-tone tasks. Smoothed cepstral peak prominence (CPPS), F0, jitter, shimmer, harmonics-to-noise ratio, corrected spectral voice-source measures, speech timing, AVQIv3-M, and time-normalized F0 contours for Tones 1-4 were analyzed. The contour root-mean-square error was used to quantify the tone stability. Generalized additive models and mixed-effects models were used to explore the age effects.Results
As age increased, CPPS, F0, harmonics-to-noise ratio, F0 range, speech rate, and maximum phonation time decreased, while jitter, shimmer, AVQIv3-M, corrected spectral measures, and CAPE-V ratings increased. The overall contour patterns of the four tones remained the same, but there were some differences in duration, excursion, and timing of these tones. Tone 3 exhibited the greatest contour variations, such as a delayed turning point and fewer measurable final rises. After adjustment for covariates, contour RMSE was positively related to age and negatively related to passage CPPS.Conclusions
In healthy aging, Mandarin-speaking women show changes both in the overall phonological quality and lexical-tone precision. The fusion of voice and tone-contour measures could help better assess Mandarin speech at the age level.