Gender Classification in Children Based on Speech Characteristics: Using Fundamental and Formant Frequencies of Malay Vowels

摘要: Speech is one of the prevalent communication mediums for humans. Identifying gender a child speaker based on his/her speech crucial in telecommunication and therapy. This article investigates use fundamental formant frequencies from sustained vowel phonation to distinguish Malay children aged between 7 12 years. The Euclidean minimum distance multilayer perceptron were used classify 360 different combinations (F0, F1, F2, F3). with normalized frequency data achieved classification accuracy 79.44%, which was higher than that nonnormalized data. Age-dependent modeling improve classification. method obtained 84.17% optimal all age groups. further increased 99.81% using mel-frequency cepstral coefficients.