How facial masks against Covid-19 spread affect the speech spectrum

Introduction: At the aim of counteracting the spread of the infection, the European Center for Disease Prevention and Control has provided indications on the suitability of the use of face masks when the social distancing cannot be implemented. Facial masks can alter verbal communication, causing potential safety problems. Methods: Eight types of masks (medical, PPE and cloth), one face-shield and four combinations of the same devices were tested, mounted on Head and Torso Simulator with an artificial mouth. For each configuration, tests were carried out in an anechoic chamber with an appropriate measurement chain for signal acquisition and processing. Results: The results show that medical devices exhibit lower attenuation at high frequencies, even compared to cloth. FFP2 and FFP3 have attenuations greater than 5 dB in the octave bands of 4 and 8 kHz, with negligible influence on acoustic performance due to the presence of valves. Face shield distorts the vocal spectrum, behaving like a low-pass filter with a cut-off frequency around 1 kHz, with a steep slope (with attenuation > 10 dB at 4 and 8 kHz) and an accentuated resonance at 1 kHz (about 6-8 dB), due to the thin PET screen. Clear mask shows low attenuation up to 4 kHz and adds the benefit of lip reading, especially useful for workers with hearing difficulties. Conclusions: These results show that the devices examined can alter the quality of speech. Further insights will concern the assessment of speech intelligibility in order to provide a contribution to an ergonomic design also linked to the acoustic comfort of these devices, which would be desirable in a post-p

Check-App Voice®: A Tool to Self-evaluate Dysphonia in Speaking Voice Among Teachers

During the SARS-Cov-2 pandemic, a tailor-made app (Check-app voice®) has been developed to submit Voice Handicap Index (VHI) questionnaire for self-assessment of dysphonia to a sample of 219 teachers. Check-app voice® provides VHI calculation and collects answers for post-processing analysis. Results have been statistically analysed taking age, gender, school categories (nursery and primary, middle, high school, and university), length of service, anamnestic data (illnesses, smoke, alcohol habits) into account. The sample of female teachers (the most numerous) has been analysed in detail. The contribution of three domains, functional, physical, and emotional to the overall VHI, has been discussed per type of school and age group. The results show that this tool represents an easy, quick to use prevention method for promptly identifying the groups most at risk of the onset of vocal disorders. © 2021, The Author(s), under exclusive license to Springer Nature Switzerland AG.