Title :
Type-2 Fuzzy GMM-UBM for Text-Independent Speaker Verification
Author :
Pinheiro, Hector N. B. ; Tsang Ing Ren ; Cavalcanti, G.D.C. ; Tsang Ing Jyh ; Sijbers, J.
Author_Institution :
Centro de Inf. (CIn), Univ. Fed. de Pernambuco (UFPE), Recife, Brazil
Abstract :
This paper proposes the use of a type-2 fuzzy framework in the standard GMM-UBM based text-independent speaker verification systems. Based on type-2 fuzzy sets, the framework provides pertinence intervals for the models. The decision process is obtained using a Support Vectors Machine (SVM) that processes the interval likelihoods. A Voice Activity Detection (VAD) algorithm was also used to discard the parts of the speech signal without voice. The proposed method was tested on the MIT Device Speaker Verification Corpus which contains several different mobile devices used in different environments. The result shows the robustness of the system and the improvements in the verification ratios of the T2F-GMM-UBM compared to the classical GMM-UBM based systems.
Keywords :
Gaussian processes; fuzzy set theory; mobile computing; speaker recognition; support vector machines; MIT device speaker verification corpus; SVM; T2F-GMM-UBM; VAD; mobile devices; support vectors machine; text-independent speaker verification; type-2 fuzzy GMM-UBM; type-2 fuzzy sets; voice activity detection algorithm; Adaptation models; Estimation; Gaussian distribution; Speech; Standards; Support vector machines; Vectors; GMM-UBM; Speaker verification; adaptative GMM-UBM; type-2 fuzzy GMM;
Conference_Titel :
Systems, Man, and Cybernetics (SMC), 2013 IEEE International Conference on
Conference_Location :
Manchester
DOI :
10.1109/SMC.2013.738