Aerosol emission in professional singing of classical music

Abstract

In this study, emission rates of aerosols emitted by professional singers were measured with a laser particle counter under cleanroom conditions. The emission rates during singing varied between 753 and 6093 particles/sec with a median of 1537 particles/sec. Emission rates for singing were compared with data for breathing and speaking. Significantly higher emission rates were found for singing. The emission enhancements between singing and speaking were between 4.0 and 99.5 with a median of 17.4, largely due to higher sound pressure levels when singing. Further, significant effects of vocal loudness were found, whereas there were no significant differences between the investigated voice classifications. The present study supports the efforts to improve the risk management in cases of possible aerogenic virus transmission, especially for choir singing.

Figure 1

Normalized emission rates for all test conditions according to the legend. Data were averaged across all eight participants and all five replicants for each test condition. Marker positions are at the mean of the particle diameter of each size class on the linear scale (see “Methods” section). Regardless of the test condition, $>99$% of all detected particles are $\le 5$ µm. The terminology, piano, mezzo-forte, and forte describes singers’ loudness conditions, namely soft, medium, and loud phonation. (Figure created with matplotlib 3.2.1, http://matplotlib.org).

Figure 2

Boxplots of the emission rates (left y-axis; bars represent the median for all replications of each task) for the test conditions breathing, speaking, and singing (denoted by hue and labeled at the bars) and the different voice classifications (baritone (red), tenor (green), alto (blue), soprano (magenta), as denoted in the x-axis) in experiment I. The emission rates for all particle size classes are cumulatively summarized. The black-framed boxes extend from the lower (Q1) to upper (Q3) quartile values of the data, with a thick line at the median. The lower and upper whiskers represent data greater than $Q1-1.5·(Q3-Q1)$ and lower than $Q3+1.5·(Q3-Q1)$, and stars denote outliers, respectively. For the test conditions speaking and singing, the median of the maximum sound pressure levels $L_{A{F}_{\mathit{MAX}}}$ for all replications were denoted by full circles (values corresponds to the right y-axis). For breathing, the sound initiated by the technical equipment such as fans was higher than any acoustic sound expelled by the participants. Therefore, no $L_{A{F}_{\mathit{MAX}}}$ are given. (Figure created with matplotlib 3.2.1, http://matplotlib.org).

Figure 3

Boxplots of the emission rates (left y-axis; bars represent the median for all replications of each task) while sustaining the vowel /aː/ in experiment II for the different vocal loudness conditions piano, mezzo-forte, and forte and voice classifications (baritone (red), tenor (green), alto (blue), soprano (magenta), as denoted in the x-axis). For the different loudness conditions, the medians of maximum sound pressure levels $L_{A{F}_{\mathit{MAX}}}$ are also shown (full circles, right y-axis). All symbols and colors correspond to the detailed description given in Fig. 2. (Figure created with matplotlib 3.2.1, http://matplotlib.org) .

Figure 4

Relationship between emission rate $P_M$ and the maximum sound pressure level $L_{A{F}_{\mathit{MAX}}}$ for the test conditions of sustained vowel /aː/ (Experiment II) for all three loudness conditions separated by voice classification including linear regression of the logarithmic emission rates (black line). For regression analysis, only medians of $P_M$ and $L_{A{F}_{\mathit{MAX}}}$ of the five replications were used. (Figure created with matplotlib 3.2.1, http://matplotlib.org) .

Figure 5

Schematic test setup with one person in cleanroom clothing whose exhaled air was recorded by the particle counter. The glass measuring section (gray colored area) was located on the suction side of a horizontally positioned Filter Fan Unit (FFU). All geometric dimensions are in mm (Figure adapted from Fig. 2 in Hartmann et al.). (Figure created with cairo 1.15.10, http://cairographics.org).