Disney Research



To date, multiple strategies have been proposed for the estimation of speakers’ physical parameters such as height, weight, age, gender etc. from their voices. These employ various types of feature measurements in conjunction with different regression and classification mechanisms. While some are quite effective for adults, they are not so for children’s voices. This is presumably because in children, the relationship between voice and physical parameters is relatively more complex. The vocal tracts of adults, and the processes that accompany speech production, are fully mature and do not undergo changes within small age differentials. In children, however, these factors change continuously with age, causing variations in style, content, enunciation, rate and quality of their speech. Strategies for the estimation of children’s physical parameters from their voice must take this variability into account. In this paper, using different formant-related measurements as exemplary analysis features generated within articulatory-phonetic guidelines, we demonstrate the nonlinear relationships of children’s physical parameters to their voice. We also show how such analysis can help us focus on the specific sounds that relate well to each parameter, which can be useful in obtaining more accurate estimates of the physical parameters.

Copyright Notice

The documents contained in these directories are included by the contributing authors as a means to ensure timely dissemination of scholarly and technical work on a non-commercial basis. Copyright and all rights therein are maintained by the authors or by other copyright holders, notwithstanding that they have offered their works here electronically. It is understood that all persons copying this information will adhere to the terms and constraints invoked by each author’s copyright. These works may not be reposted without the explicit permission of the copyright holder.