Abstract
In this investigation the voice source from trained boy singers was processed with a transfer function that contained the singer's formant cluster of a bass, a baritone, or a tenor. The modified voices were evaluated by a panel of highly specialized experts. The experts were asked 1) to assess how similar the examples sounded to the voice of the last castrato Alessandro Moreschi, and 2) to rate how similar they thought the examples were to their imagination of an 18th-century castrato voice. For both questions, the voices with tenor formants produced significantly higher ratings than the other voice types. However, the mean ratings for the second question were generally lower than those for the first.
Acknowledgements
This investigation was originally presented at the 3rd International Conference on the Physiology and Acoustics of Singing (PAS) in York, 2006. The authors are indebted to the boys of the Tölzer Knabenchor and to their voice teachers and their director, Gerhard Schmidt-Gaden, for their kind and helpful co-operation. Also, the patient help of the experts participating in the listening panel is gratefully acknowledged.
Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.