Classifying Singing Voices

Version française...

Voice classification is a frequent preoccupation of young singers searching for repertoire. Often, singers are sent to a laryngologist in the hope that the anatomy of their vocal folds and throat will help in their voice classification. The prevailing misconception among novices is that vocal fold length determines voice type. However, classification must also take into account voice quality and pitch (perceived frequency) range, and not just the fundamental frequency produced by the vibrating vocal folds.

In general, longer vocal folds in males (18 to 24 mm) than in females (14 to 19 mm) correspond to deeper voices. Determining vocal fold length by direct examination is not an easy task. One must consider only the vibrating membranous part of the vocal fold and account for variations in length due to vocal muscle contraction. Varying vocal fold length is not the only factor determining higher fundamental frequency in singing; changes in vocal muscle tension are also involved.

The shape and volume of the resonating oral cavities are also important since they give the colour to the voice. Theoretically, for a given vocal fold length, the voice is perceived as darker (lower) if the vocal tract is longer. To a certain extent, singers can control the length of their resonating vocal tracts by raising or lowering their larynx or pursing their lips.

Hence, voice classification must be based on dynamic physiological parameters. There is currently no rigorous scientific method to determine voice classification. However, there is agreement on the following criteria:

1) The tessitura. This is the part of the vocal range which the singer is most comfortable producing: for tenors, C3 to C5; for baritones, A2 to A4; for basses, E2 to E4; for sopranos, C4 to C6; for mezzo-sopranos, A3 to A5; for altos, E3 to G5.

2) The passaggio. This is the range of frequencies at which a register change occurs: for tenors, E4 to F4; for baritones, D4 to E4; for basses, C4 to D4. The passaggio for female voices is one octave higher than in the male.

3) The overall spectrum of acoustic energy in the voice. Vocal sound contains frequencies generated by the vocal folds that are shaped into peaks (formants) by the resonating vocal tract. Cleveland (1978) proposed a classification of the male singing voice based on average spectral peaks of acoustic energy extracted from the singing of an extended passage. Objective analysis of the acoustic energy peaks allows for hybrid classifications that occur when short vocal cords are coupled with a long vocal tract, or vice versa.

The objective of classifying the voice is to steer singers towards a repertoire that best suits their anatomical and physiological potential. There is no need for premature classification, as the inexperienced singer's tessitura and vocal range are underdeveloped.

Voice classification is best done by the vocal pedagogue. Laryngologists may assist by correcting impeding vocal fold pathologies. Acoustical analysis serves as a pedagogical aid by highlighting the importance of modulation of the vocal tract resonators in formant development and enrichment of the acoustical output. Ultimately, proper voice classification follows from optimizing the singer's physical reality through correct vocal training.