Speaker weight estimation from speech signals using a fusion of the i-vector and NFA frameworks

In this paper, a novel approach for automatic speaker weight estimation from spontaneous telephone speech signals is proposed. In this method, each utterance is modeled using the i-vector framework which is based on the factor analysis on Gaussian Mixture Model (GMM) mean supervectors, and the Non-negative Factor Analysis (NFA) framework which is based on a constrained factor analysis on GMM weights. Then, the available information in both Gaussian means and Gaussian weights is exploited through a feature-level fusion of the i-vectors and the NFA vectors. Finally, a least-squares support vector regression (LS-SVR) is employed to estimate the weight of speakers from given utterances. The proposed approach is evaluated on the telephone speech signals of National Institute of Standards and Technology (NIST) 2008 and 2010 Speaker Recognition Evaluation (SRE) corpora. Experimental results over 2339 utterances show that the correlation coefficients between actual and estimated weights of male and female speakers are 0.56 and 0.49, respectively, which indicate the effectiveness of the proposed method in speaker weight estimation.