Maximum Likelihood and Maximum Mutual Information Training in Gender and Age Recognition System

Abstract. Gender and age estimation based on Gaussian Mixture Models (GMM) is introduced. Telephone recordings from the Czech SpeechDatEast database are used as training and test data set. Mel-Frequency Cepstral Coefficients (MFCC) are extracted from the speech recordings. To estimate the GMMs’ parameters Maximum Likelihood (ML) training is applied. Consequently these estimations are used as the baseline for Maximum Mutual Information (MMI) training. Results achieved when employing both ML and MMI training are presented and discussed.

• Type: Conference
• Year: 2007
• Where: TSD
• Authors: Valiantsina Hubeika, Igor Szöke, Lukas Burget, Jan Cernocký
• Keywords: Gaussian Mixture Models, Mel-Frequency Cepstral Coefficients, Parameters Maximum Likelihood, Signal Processing, TSD 2007