Objective: To investigate effects of surgical and transparent face masks during the Covid-19 pandemic on audiovisual speech recognition of words for deaf and hard-of-hearing children.

Design: Recorded Word Intelligibility by Picture Identification test (WIPI) was presented in quiet via a computer monitor to children in a quiet test room. The acoustic power spectra of each mask type were compared to the baseline no mask condition. Percent correct word recognition was recorded for four mask conditions (no mask, surgical mask, transparent apron mask and ClearMask) in counterbalanced order. Repeated measures ANOVA was used to test for significant differences in word recognition scores across mask types.

Study Sample: Thirteen children (3 to 7 years) in a private auditory oral school wearing hearing aids, bone-anchored hearing aids or cochlear implants. Children were excluded if English was not their primary language or if they had a severe speech-language delay, uncorrected vision loss, or developmental disorder that would affect the results. No children had been exposed to or had contracted the Covid-19 virus.

Results: Acoustic spectra showed a decrease in the 2000-8000 Hz region for the transparent apron mask. The surgical mask and ClearMask showed fewer acoustic effects. Children with hearing aids performed similarly to children with cochlear implants. Word recognition was significantly poorer for surgical masks and transparent apron masks. The ClearMask condition was not significantly worse than the no mask condition for words in quiet.

Conclusions: Standard surgical and custom apron shield masks significantly hampered word recognition, even in quiet conditions. The commercially available ClearMask did not significantly affect scores in quiet for young deaf and hard-of-hearing children, but scores were highly variable.