Key points are not available for this paper at this time.
Children, aged 3–8 years, speaking different native languages, were recorded and speech samples from these children were played back to adult listener groups, also of different language background. The listeners were asked to judge the sex of the child. The children were native speakers of English, Finnish, or Swedish and the listeners were native speakers of Finnish, Swedish, English, or Chinese. Speakers and listeners from different language backgrounds were used to find out if the differences in boys' and girls' voices were socially acquired or innate. The average right judgment was 66% over all speakers and listeners, with no strong tendency for a higher percentage of right answers for children speaking the listener's own language. The results from the listening tests have been compared to different acoustic measurements of F0, vowel formants, and speaking rate. No apparent correlation between perceived sex and these measurements has been found.
Karlsson et al. (Fri,) studied this question.