Acoustic characteristics of perceived versus actual vocal age in controlled phonation by adult females

TLDR

The study aimed to characterize how acoustic features of women’s voices differ between perceived and actual age. The authors recorded whispered and normally phonated /ae/ vowels from 75 women in three age groups, had listeners judge age from 1‑second excerpts, and analyzed phonated vowels for F0, F0 SD, jitter, F1, and F2 and whispered vowels for F1 and F2. Older age judgments from phonated vowels correlated with higher F0 variability and lower mean F0, while judgments from whispered vowels correlated with lower F1; overall, both phonatory and resonance cues align with perceived age and reflect actual age differences.

Abstract

The purpose of this investigation was to obtain information on the acoustic characteristics of women's voices as a function of perceived versus actual age. Tape recordings of whispered and normally phonated /ae/ vowels were obtained from 75 women in three age groups (25-35, 45-55, and 70-80 years). One-second segments of phonemically representative /ae/ vowels were played to 23 listeners for relative age judgments. Phonated vowel productions were analyzed in terms of mean fundamental frequency, fundamental frequency standard deviation, and jitter ratio, as well as F1 and F2 frequencies. Whispered vowels were analyzed only in terms of F1 and F2 frequencies. Multiple regression analysis revealed that older age judgments based on phonated vowels were significantly associated with higher fundamental frequency standard deviation values and lower mean fundamental frequency values. Older age judgments based on whispered vowels were significantly associated with lower F1 frequencies. Results of this investigation suggest that both phonatory and resonance features play a role in defining age characteristics of women's voices. Acoustic cues associated with perceived age generally agreed with acoustic findings as a function of actual age.