Purpose: This study tested whether dysphonic voice quality reduces vowel recognition in noise in school-age children compared to that in adults. We also examined whether perceptual and acoustic indices of dysphonia, which included Consensus Auditory-Perceptual Evaluation of Voice (CAPE-V) ratings and cepstral peak prominence (CPP), could predict recognition accuracy across listeners. Method: Participants were 26 children (aged 8–17 years, M = 12.8) and 13 adults (aged 18–21 years, M = 20.6), all native English speakers with normal hearing. Listeners identified 11 American English vowels in an /hVd/ context, produced by three female talkers with typical voices and three with dysphonia, confirmed by laryngologist diagnosis. Stimuli were presented in speech-shaped noise matched to each talker's long-term spectrum at −6 dB and −9 dB signal-to-noise ratio (SNR). Each vowel was presented three times per condition, once per talker, yielding equal numbers of trials across voice quality and SNR. CAPE-V ratings of overall severity and strain were obtained from expert listeners, and CPP was extracted from vowel tokens using established acoustic procedures. Results: Recognition accuracy was reduced at poorer SNRs and dysphonia compared with typical voices. Overall, adults outperformed children. In the child group, recognition improved with age, but this developmental advantage was attenuated for dysphonic voices. Greater CAPE-V severity and strain ratings predicted poorer recognition across groups, and higher CPP values were associated with better performance. Conclusions: Dysphonic voice quality significantly constrains children's vowel recognition in noise, consistent with reduced access to source-related acoustic detail that supports developmental improvement. These results highlight the combined importance of teachers' vocal health and classroom acoustics in supporting children's speech perception and learning.
Flaherty et al. (Mon,) studied this question.