Gilles Boulianne

Joint Factor Analysis Versus Eigenchannels in Speaker Recognition

We compare two approaches to the problem of session variability in Gaussian mixture model (GMM)-based speaker verification, eigenchannels, and joint factor analysis, on the National Institute of Standards and Technology (NIST) 2005 speaker recognition evaluation data. We show how the two approaches can be implemented using essentially the same software at all stages except for the enrollment of target speakers. We demonstrate the effectiveness of zt-norm score normalization and a new decision criterion for speaker recognition which can handle large numbers of t-norm speakers and large numbers of speaker factors at little computational cost. We found that factor analysis was far more effective than eigenchannel modeling. The best result we obtained was a detection cost of 0.016 on the core condition (all trials) of the evaluation

Eigenvoice modeling with sparse training data

We derive an exact solution to the problem of maximum likelihood estimation of the supervector covariance matrix used in extended MAP (or EMAP) speaker adaptation and show how it can be regarded as a new method of eigenvoice estimation. Unlike other approaches to the problem of estimating eigenvoices in situations where speaker-dependent training is not feasible, our method enables us to estimate as many eigenvoices from a given training set as there are training speakers. In the limit as the amount of training data for each speaker tends to infinity, it is equivalent to cluster adaptive training.

Speaker and Session Variability in GMM-Based Speaker Verification

We present a corpus-based approach to speaker verification in which maximum-likelihood II criteria are used to train a large-scale generative model of speaker and session variability which we call joint factor analysis. Enrolling a target speaker consists in calculating the posterior distribution of the hidden variables in the factor analysis model and verification tests are conducted using a new type of likelihood II ratio statistic. Using the NIST 1999 and 2000 speaker recognition evaluation data sets, we show that the effectiveness of this approach depends on the availability of a training corpus which is well matched with the evaluation set used for testing. Experiments on the NIST 1999 evaluation set using a mismatched corpus to train factor analysis models did not result in any improvement over standard methods, but we found that, even with this type of mismatch, feature warping performs extremely well in conjunction with the factor analysis model, and this enabled us to obtain very good results (equal error rates of about 6.2%)

Factor analysis simplified [speaker verification applications]

We show how the factor analysis model for speaker verification can be successfully implemented using some fast approximations which result in minor degradations in accuracy and open up the possibility of training the model on very large databases such as the union of all of the Switchboard corpora. We tested our algorithms on the NIST 1999 evaluation set (carbon data as well as electret). Using warped cepstral features we obtained equal error rates of about 6.3% and minimum detection costs of about 0.022.

Generating exact lattices in the WFST framework

We describe a lattice generation method that is exact, i.e. it satisfies all the natural properties we would want from a lattice of alternative transcriptions of an utterance. This method does not introduce substantial overhead above one-best decoding. Our method is most directly applicable when using WFST decoders where the WFST is “fully expanded”, i.e. where the arcs correspond to HMM transitions. It outputs lattices that include HMM-state-level alignments as well as word labels. The general idea is to create a state-level lattice during decoding, and to do a special form of determinization that retains only the best-scoring path for each word sequence. This special determinization algorithm is a solution to the following problem: Given a WFST A, compute a WFST B that, for each input-symbol-sequence of A, contains just the lowest-cost path through A.

Combining Gaussianized/Non-Gaussianized Features to Improve Speaker Diarization of Telephone Conversations

We report results on speaker diarization of telephone conversations. This speaker diarization process is similar to the multistage segmentation and clustering system used in broadcast news. It consists of an initial acoustic change point detection algorithm, iterative Viterbi re-segmentation, gender labeling, agglomerative clustering using a Bayesian information criterion (BIC), followed by agglomerative clustering using state-of-the-art speaker identification (SID) methods and Viterbi re-segmentation using Gaussian mixture models (GMMs). We repeat these multistage segmentation and clustering steps twice: once with mel-frequency cepstral coefficients (MFCCs) as feature parameters for the GMMs used in gender labeling, SID, and Viterbi re-segmentation steps and another time with Gaussianized MFCCs as feature parameters for the GMMs used in these three steps. The resulting clusters from the parallel runs are combined in a novel way that leads to a significant reduction in the diarization error rate (DER). On a development set containing 30 telephone conversations, this combination step reduced the DER by 20%. On another test set containing 30 telephone conversations, this step reduced the DER by 13%. The best error rate we have achieved is 6.7% on the development set and 9.0% on the test set.

Improvements in Factor Analysis Based Speaker Verification

We present the results of speaker verification experiments conducted on the NIST 2005 evaluation data using a factor analysis of speaker and session variability in 6 telephone speech corpora distributed by the Linguistic Data Consortium. We demonstrate the effectiveness of zt-norm score normalization and a new decision criterion for speaker recognition which can handle large numbers of t-norm speakers and large numbers of speaker factors at little computational cost. The best result we obtained was a detection cost of 0.016 on the core condition (all trials) of the evaluation

Speaker adaptation using an eigenphone basis

We describe a new method of estimating speaker-dependent hidden Markov models for speakers in a closed population. Our method differs from previous approaches in that it is based on an explicit model of the correlations between all of the speakers in the population, the idea being that if there is not enough data to estimate a Gaussian mean vector for a given speaker then data from other speakers can be used provided that we know how the speakers are correlated with each other. We explain how to estimate inter-speaker correlations using a Kullback-Leibler divergence minimization technique which can be applied to the problem of estimating the parameters of all of the hyperdistributions that are currently used in Bayesian speaker adaptation.

Speaker diarization of French broadcast news

We report results on speaker diarization of French broadcast news and talk shows on current affairs. This speaker diarization process is a multistage segmentation and clustering system. One of the stages is agglomerative clustering using state-of-the-art speaker identification methods (SID). For the QMMs used in this stage, we tried many different feature parameters, including MFCCs, Gaussianized MFCCs, Gaussianized MFCCs with cepstral mean subtraction, and Gaussianized MFCCs with cepstral mean substraction containing only frames with high energy. We found that this last set of feature parameters gave the best results. Compared to Gaussianized MFCCs, these features reduced the diarization error rate (DER) by 12% on a development set and by 19% on a test set. We also combined clusters resulting from Gaussianized and non-Gaussianized feature sets. This cluster combination resulted in another 4% reduction in DER for both the development and the test sets. The best DER we have achieved is 15.4% on the development set, and 14.5% on the test set.

Optimal tying of HMM mixture densities using decision trees

The most detailed acoustic models in our two-pass speaker-independent, continuous speech recognition system are context-dependent models, which become more difficult to adequately train as the number of different contexts becomes large. Tying of model parameters or clustering of model densities based on bottom-up agglomerative procedures can efficiently reduce the number of parameters to train, but suffer from the additional problem of how to model untrained contexts. Top-down clustering with a decision tree can provide well-trained models for any context, whether seen or unseen in training. Trees are built from a root node that is successively split by selecting, among questions about phonetic context, one that provides the best segregation of data. Several goodness of split criterions have been proposed, such as Poisson-based (Bahl et al., 1991), or single Gaussian-based (Bahl et al., 1994), their choice being primarily motivated by computational considerations. We show, from maximum likelihood considerations, how to derive a computationally efficient criterion based on a different approximation using tied mixtures of Gaussian densities.