Music performers convey different emotions to their audience through their performance. Previous research has revealed acoustical performance parameters differentiating basic emotions (e.g., happiness, sadness, anger, fear). However, realistic situations like opera often feature scenarios with mixed emotions, such as “love and hate” or “courage and fear.” We have previously reported that listeners successfully differentiated the singers’ intended levels of emotions across five happiness–sadness ratios: 100% sadness, 70% sadness/30% happiness, 50% sadness/50% happiness, 30% sadness/70% happiness, and 100% happiness. In this study, we examined how these five renditions and two musical works (major versus minor) influenced the acoustic characteristics of the singing voice. We acoustically analyzed the recordings of 12 expert vocalists (six men and six women) singing five renditions (emotional ratios: 10:0, 7:3, 5:5, 3:7, and 0:10) each of Franz Wüllner’s “Chorübungen” No. 41a (major) and No. 80c (minor). Results showed that the spectral centroid as well as the first (F1) and second formants (F2) depended on the emotional ratios. These findings suggest that singers manipulate their mouth openness (F1) and tongue position (F2) to differentiate timbre brightness and thus achieve their intended levels of happiness/sadness.
Shoda et al. (Wed,) studied this question.