Utilização da técnica MFCC em conjunto com os parâmetros extraídos do sinal glotal para melhorar o desempenho de um sistema de verificação de locutor

This article presents the development of a text dependent speaker verification algorithm, which alias the Mel Frequency Cepstrum Coeficients with parameters calculated from the glottal signal using Hidden Markov Models to identify the stochastic pattern of these measurements. The glottal signal is generated when the air flow passes through the vocal chord, and is obtained in this work by inverse filtering.