Measuring the Differences Between Opera and Musical Theatre Styles Through Spectral Moment Analysis

Measuring Differences Between Opera and Musical Theatre Singing Through Spectral Moment Analysis

Objective: Established spectral analysis methods, including LTAS and formant tuning, have been used to describe acoustic differences in various genres of singing. However, further research is required to compare acoustic differences between genres systematically. Spectral moments have been used to describe the EGG decontacting pattern, recording technology, measures of patients with dysphonia, speech sounds, and synthesized singing with individually altered fo, vibrato rate, vibrato extent, vowel (fR1, fR2, and fR3), spectral tilt, and duration. This study will expand on previous findings by exploring spectral moments as a means to characterize genre differences in the singing voice.

Method: Four cis-gender female soprano singers with professional performance experience in both musical theatre and opera will sing English-language excerpts from standard repertoire. Calibrated voice recordings will be taken with an Earthworks M23 flat response, omnidirectional cardioid microphone in a controlled acoustic environment. Each excerpt contains a sustained pitch/vowel of E4 on [I], Bb4 on [ɔ], or E5 on [ε]. LTAS from genre-paired pitch/vowels will be analyzed using spectral moments in PRAAT. Spectral mean (M1), spectral standard deviation (M2), spectral skewness (M3), and spectral kurtosis (M4) will be calculated for each audio sample. Linear models with modified Bonferroni corrections will be applied.

Results/Conclusion: Data collection and analysis are ongoing.