Search

This paper studies the relevance factor in maximum a posteriori (MAP) adaptation of Gaussian mixture model (GMM) for speaker and language recognition. Knowing that relevance factor determines how much the observed training data influence the model adaptation, thus the resulting GMM model, it is believed that more effective modeling can be achieved if the relevance factor is adaptive to the corresponding data. We therefore provide a mathematic derivation for the estimation of relevance factor. GMM supervector support vector machine (SVM) with nuisance attribute projection (NAP) (GMM-NAP-SVM) has been reported to be effective and reliable for speaker and language recognition. Being a discriminative classifier in nature, a GMM-NAP-SVM system is sensitive to the magnitude and direction of a supervector in the high dimensional space. However, when characterizing a speech utterance with GMM supervector estimated through MAP, we observe that the resulting supervector is undesirably affected by the varying duration of the utterance. We propose an adaptive relevance factor that adapts to the duration to mitigate the variability effect due to the length of utterance. We give a systematic investigation on different types of relevance factor of MAP in different applicatively platforms. We show the efficacy of the data-dependent as well as adaptive relevance factors on the National Institute of Standards and Technology (NIST) speaker recognition evaluation (SRE) 2008 and language recognition evaluation (LRE) 2009 and 2011 tasks respectively.

Support vector machines (SVMs), and kernel classifiers in general, rely on the kernel functions to measure the pairwise similarity between inputs. This paper advocates the use of discrete representation of speech signals in terms of the probabilities of discrete events as feature for speaker verification and proposes the use of Bhattacharyya coefficient as the similarity measure for this type of inputs to SVM. We analyze the effectiveness of the Bhattacharyya measure from the perspective of feature normalization and distribution warping in the SVM feature space. Experiments conducted on the NIST 2006 speaker verification task indicate that the Bhattacharyya measure outperforms the Fisher kernel, term frequency log-likelihood ratio (TFLLR) scaling, and rank normalization reported earlier in literature. Moreover, the Bhattacharyya measure is computed using a data-independent square-root operation instead of data-driven normalization, which simplifies the implementation. The effectiveness of the Bhattacharyya measure becomes more apparent when channel compensation is applied at the model and score levels. The performance of the proposed method is close to that of the popular GMM supervector with a small margin.

International audience; This paper advocates the use of probabilistic linear discriminant analysis (PLDA) for partially open-set detection task with multiple i-vectors enrollment condition. Also referred to as speaker verification, the speaker detection task has always been considered under an open-set scenario. In this paper, a more general partially open-set speaker detection problem in considered, where the imposters might be one of the known speakers previously enrolled to the system. We show how this could be coped with by modifying the definition of the alternative hypothesis in the PLDA scoring function. We also look into the impact of the conditional-independent assumption as it was used to derive the PLDA scoring function with multiple training i-vectors. Experiments were conducted using the NIST 2012 Speaker Recognition Evaluation (SRE'12) datasets to validate various points discussed in the paper.

In this paper, we propose a self-organized clustering method for feature mapping to compensate the channel variation in spoken language recognition. The self-organized clustering is realized by transforming the utterances into the Gaussian mixture model (GMM) supervectors and categorizing the supervectors through k-mean algorithm. Based on the language-dependent cluster-of-utterance information of the training databases, the feature mapping parameters are trained for each of the target languages. During recognition, the test utterance is identified to be one of the clusters according to the feature mapping parameters and then transformed into the cluster-independent features through feature mapping for a given target language. We show the effectiveness of the proposed self-organized feature mapping scheme through the 2003 National Institute of Standards and Technology (NIST) Language Recognition Evaluation (LRE) by using GMM recognizer.

This thesis focuses on the development of robust speech enhancement methods based on masking properties of the human auditory system. DOCTOR OF PHILOSOPHY (EEE)