Evaluation of Puberty in Girls by Spectral Analysis of Voice

In this paper, a method for girls’ pubertal status evaluation is presented. The proposed algorithm uses voice features. Spectral analysis, Support Vector Machine and Random Forest Trees were employed. The obtained results are promising. Sensitivity reached 89.38%, when all features were included in the calculations (SVM). The highest specificity was achieved when only standard deviations were used (80.14% for the RF). Accuracy was greater than 80% for both classifiers when all features were used.

[1]  Jont B. Allen,et al.  Short term spectral analysis, synthesis, and modification by discrete Fourier transform , 1977 .

[2]  Alexander J. Smola,et al.  Learning with Kernels: support vector machines, regularization, optimization, and beyond , 2001, Adaptive computation and machine learning series.

[3]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[4]  Joel MacAuslan,et al.  Toward clinical application of landmark-based speech analysis: Landmark expression in normal adult speech. , 2017, The Journal of the Acoustical Society of America.

[5]  S. Hawkins,et al.  Changes in the male voice at puberty , 1997, Archives of disease in childhood.

[6]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[7]  John H. L. Hansen,et al.  Unsupervised Speech Activity Detection Using Voicing Measures and Perceptual Spectral Flux , 2013, IEEE Signal Processing Letters.

[8]  M. B. Manjunatha,et al.  Speech Features Extraction Techniques for Robust Emotional Speech Analysis/Recognition , 2017 .

[9]  Ingo R. Titze,et al.  Principles of voice production , 1994 .

[10]  Eric Granger,et al.  Feature Learning from Spectrograms for Assessment of Personality Traits , 2016, IEEE Transactions on Affective Computing.

[11]  D. Perrett,et al.  Manipulations of fundamental and formant frequencies influence the attractiveness of human male voices , 2005, Animal Behaviour.

[12]  Tin Kam Ho,et al.  A Data Complexity Analysis of Comparative Advantages of Decision Forest Constructors , 2002, Pattern Analysis & Applications.

[13]  Marcin D. Bugdol,et al.  Prediction of menarcheal status of girls using voice features , 2017, Comput. Biol. Medicine.

[14]  Ieuan A. Hughes,et al.  A wider perspective on puberty , 2006, Molecular and Cellular Endocrinology.

[15]  S. Cunningham,et al.  Analyses of Sustained Vowels in Down Syndrome (DS): A Case Study Using Spectrograms and Perturbation Data to Investigate Voice Quality in Four Adults With DS. , 2017, Journal of voice : official journal of the Voice Foundation.

[16]  K. Pisanski,et al.  Seven and up: individual differences in male voice fundamental frequency emerge before puberty and remain stable throughout adulthood , 2016, Royal Society Open Science.

[17]  J. Perheentupa,et al.  Fundamental voice frequence during normal and abnormal growth, and after androgen treatment. , 1978, Archives of disease in childhood.

[18]  M. Gurven,et al.  Voice change as a new measure of male pubertal timing: A study among Bolivian adolescents , 2013, Annals of human biology.

[19]  Olga Karapanou,et al.  Determinants of menarche , 2010, Reproductive biology and endocrinology : RB&E.