Research project: Gender in Children's Voices

Dissertation project by: Riccarda Funk

Project leader: Adrian P. Simpson

SI 743/9-1/2, Duration May 2020 to November 2027

Until puberty, there are hardly any anatomical and physiological differences in the vocal tract of girls and boys. Nevertheless, gender-specific differences in prepubertal voices have been repeatedly found, sometimes as early as the age of two and a half years. Furthermore, adult listeners are able to correctly identify a child's gender from speech samples at levels above chance. In some children's voices, recognition even exceeds 90%, reaching recognition rates comparable to those found in adult voices.

This project investigates the perceptual and acoustic correlates of gender in prepubertal voices through a longitudinal study. To date, speech recordings of the same children have been made at five time points (from first to fifth grade). Since the children's gender conformity may play an important role in acoustic and perceptual variation, this will also be assessed as part of the project. Additionally, various listening experiments will be conducted.

The aim of the project is to identify perceptual parameters that lead to robust or ambivalent gender assignments. We will investigate if these parameters correlate with a child's gender conformity. By conducting this as a longitudinal study, changes in the acoustic correlates of gender and gender conformity throughout childhood can also be observed.

All recordings made in the project will be collected in the LoKiS corpus. Interested researchers may be granted access to these data upon request.