Speaking fundamental frequency in (pre)pubertal female choristers: A longitudinal study

Humans exhibit vocal sexual dimorphism driven by hormonally mediated laryngeal changes during puberty, yet longitudinal evidence for female speaking fundamental frequency (SFF) development remains limited. This study draws on an unprecedented longitudinal corpus of 75 female choristers (aged 8-18 years; mean: 12.70 ± 2.15) recorded at Wells Cathedral School (UK) at six-month intervals between 1999 and 2017, averaging 8.12 sessions per participant and yielding 609 acquisitions in total. Participants read a phonetically balanced passage at conversational intensity; SFF was estimated via Praat's autocorrelation algorithm and modeled using a linear mixed-effects framework. The model revealed significant effects of age, quadratic age, cubic age, and starting age (age of choir entry), with all p values ≤0.019 and no significant interactions. SFF declined non-linearly with age, plateauing around 17 years. Later choir entry was associated with higher SFF, suggesting that earlier or more extensive vocal training modestly elevates SFF overall. The model explained 30.29% of total variance, with significant random effects reflecting considerable inter-individual variability in timing and magnitude of pubertal SFF change. These findings underscore the multifactorial nature of female voice development and suggest that sparse temporal sampling may insufficiently capture the complexity of maturational trajectories.