Speaker verification using Gaussian Mixture Model

In this paper, speaker verification system using Gaussian Mixture Model (GMM) is proposed. The proposed system consists of pre-processing, feature extraction, modelling and classification stage. The pre-processing is used to remove silent part of signal to reduce computational complexity. Pitch frequency and Mel Frequency Cepstral Coefficients(MFCC)are used as a feature vector for speaker verification system. Modelling is done using different combination of Gaussian mixture models. Simple distance measures are used for the classification between reference and the test signal.

[1]  S. R. M. Prasanna,et al.  Significance of Vowel-Like Regions for Speaker Verification Under Degraded Conditions , 2011, IEEE Transactions on Audio, Speech, and Language Processing.

[2]  William M. Campbell,et al.  Speaker Verification Using Support Vector Machines and High-Level Features , 2007, IEEE Transactions on Audio, Speech, and Language Processing.

[3]  Sridha Sridharan,et al.  Making Confident Speaker Verification Decisions With Minimal Speech , 2010, IEEE Transactions on Audio, Speech, and Language Processing.

[4]  Haizhou Li,et al.  Low-Variance Multitaper MFCC Features: A Case Study in Robust Speaker Verification , 2012, IEEE Transactions on Audio, Speech, and Language Processing.

[5]  S R M Prasanna,et al.  Multi-variability speech database for robust speaker recognition , 2011, 2011 National Conference on Communications (NCC).

[6]  Bayya Yegnanarayana,et al.  Combining evidence from residual phase and MFCC features for speaker recognition , 2006, IEEE Signal Processing Letters.

[7]  Longbiao Wang,et al.  Speaker Identification and Verification by Combining MFCC and Phase Information , 2009, IEEE Transactions on Audio, Speech, and Language Processing.

[8]  S. R. Mahadeva Prasanna,et al.  Combining evidence from source, suprasegmental and spectral features for a fixed-text speaker verification system , 2005, IEEE Transactions on Speech and Audio Processing.