Male/Female Speech Classification Based On Cepstral Modulation Ratio Parameterization by Laguerre Polynomials
Keywords:
Gender Classification, Cepstral ModulationParameters, Laguerre Polynomials.
Abstract:
This paper uses a new set of feature vectors that isbased on modulation spectrum of cepstral coefficients by meansof Laguerre regression method. The performance of theproposed method is investigated by a gender classification of anoisy speech. Compared with other regression methods, ourproposed feature set demonstrates high performance in thesense of gender classification. Low classification errorsobtained in different noisy scenarios proves the superiority ofthe new feature vectors for the classification task.