Fig. 1
From: Effects of experience on recognition of speech produced with a face mask
Average performance by talker, mask, and SNR from the current dataset (collected in 2021). Similar to Toscano and Toscano (2021), listeners were more accurate in the high SNR condition, and more accurate for Talker 1. Compared to the no-mask condition, the surgical mask yielded the best performance, followed by the N95 respirator, and the two cloth masks. Dashed lines indicate the average performance for the no-mask condition in each panel. Error bars indicate 95% confidence intervals