Aerosol emissions during speech and whispering among 41 professional and semi-professional singers

The aerosol transmission of pathogens has gathered interest since the COVID-19 pandemic, but it seems to be important for many upper respiratory tract infections. The aim of this study was to assess how different vocalization affects aerosol emissions. 41 experienced singers repeated two different sentences in normal loudness they would use talking at coffee room and whispering. The sentences had different phonations with sentence 1 including mainly consonants /k/ and /t/ while sentence 2 also had /s/. The aerosol emission was measured using condensation particle counter (CPC 3775, TSI Inc.) and aerodynamic particle sizer (APS 3321, TSI Inc.). APS data was divided to > 1 µm and < 1 µm particle sizes. CPC measures the concentration of particles > 4 nm and reflects the number of the smallest particles. Mann-Whitney U test was used for significance testing with Benjamini-Hochberg correction for multiple comparisons. The figure shows aerosol emissions for each sentence and vocalization. Whispering produced more aerosols than speaking and sentence 1 produced more aerosols than sentence 2 while speaking. The differences in exhaled flow were not statistically significant and differences between whispering were significant only for the largest particles. Earlier sound pressure has been identified as the most important factor associated with higher aerosol emissions. Based on our results vocalization also has a significant effect on particle production.