Abstract
Face masks are an important tool for preventing the spread of COVID-19. However, it is unclear how different types of masks affect speech recognition in different levels of background noise. To address this, we investigated the effects of four masks (a surgical mask, N95 respirator, and two cloth masks) on recognition of spoken sentences in multi-talker babble. In low levels of background noise, masks had little to no effect, with no more than a 5.5% decrease in mean accuracy compared to a no-mask condition. In high levels of noise, mean accuracy was 2.8-18.2% lower than the no-mask condition, but the surgical mask continued to show no significant difference. The results demonstrate that different types of masks generally yield similar accuracy in low levels of background noise, but differences between masks become more apparent in high levels of noise.
Highlights
Human speech perception is remarkably robust across a wide range of contexts
We examined the effects of two homemade cloth masks, a surgical mask, and an N95 respirator, comparing performance to speech produced without a mask under conditions of both high and low levels of background noise
The results demonstrate that masks affect speech recognition to varying degrees depending on the talker and level of background noise
Summary
Human speech perception is remarkably robust across a wide range of contexts. Listeners with normal hearing can understand speech even in relatively high levels of background noise [1], and they can cope with considerable acoustic variability between talkers’ voices [2, 3]. A potential concern with the use of face masks is that they may cause a reduction in speech intelligibility We addressed this concern by investigating the effects of different types of masks on speech recognition in the context of multi-talker babble noise. The specific context evaluated here involves conditions in which no visual cues are available, with recordings made using a microphone at close distance and sentences normalized to have the same average intensity This differs from in-person face-to-face communication in important ways, it is directly applicable to contexts in which a talker may need to use a microphone while wearing a mask (e.g., while teaching), and it provides information about how the type of mask and level of background noise affect speech communication. We measured the proportion of words in each sentence that listeners correctly recognized to assess the effect of each type of mask on speech recognition
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
