A comparative analysis of the singer's formant cluster

[摘要] ENGLISH ABSTRACT: It is widely accepted that the singer's formant cluster (Fs) – perceptual correlates being twangand ring, and pedagogically referred to as head resonance – is the defining trait of aclassically trained voice. Research has shown that the spectral energy a singer harnesses inthe Fs region can be measured quantitatively using spectral indicators Short-Term EnergyRatio (STER) and Singing Power Ratio (SPR). STER is a modified version of the standardmeasurement tool Energy Ratio (ER) that repudiates dependency on the Long-Term AverageSpectrum (LTAS). Previous studies have shown that professional singers produce more Fsspectral energy when singing in ensemble mode than in solo mode; however for amateursingers, the opposite trend was noticed. Little empirical evidence in this regard is availableconcerning undergraduate vocal performance majors. This study was aimed at investigatingthe resonance tendencies of individuals from the latter target group, as evidenced whensinging in two performance modes: ensemble and solo. Eight voice students (two per SATBvoice part) were selected to participate. Subjects were recorded singing their partsindividually, as well as in full ensemble. By mixing the solo recordings together, comparisonsof the spectral content could be drawn between the solo and ensemble performance modes.Samples (n=4) were extracted from each piece for spectral analyses. STER and SPR meanswere highly proportional for both pieces. Results indicate that the singers producesignificantly higher levels of spectral energy in the Fs region in ensemble mode than in solomode for one piece (p<0.05), whereas findings for the other piece were insignificant. Thefindings of this study could inform the pedagogical approach to voice-training, and providesempirical bases for discussions about voice students' participation in ensemble ventures.

[发布机构] Stellenbosch University