A Gaussian, missing-data model is applied to predict product ratings. Vectors of product ratings from users are assumed to be independent and identically distributed. Two approaches for parameter estimation in this model are studied: Little and Rubin's expectation-maximization algorithm and McMichael's modified stochastic gradient descent approach. The resulting estimates are used in minimum mean squared error prediction of product ratings using the conditional mean. On a large dataset, performance using McMichael's approach is better than reported performance of the popular matrix factorization approach.

Recursive Markov-modulated Poisson process (MMPP) parameter estimation is performed by adapting an approach for hidden Markov model estimation developed by Krishnamurthy and Moore. Explicit expressions are developed for functions used in the recursion. The resulting approach is compared to a recursive MMPP estimation algorithm developed by Lindgren and Holst. Numerical results are provided which demonstrate the applicability of the approach for estimation of interrupted Poisson processes.

Two aspects of Ryden's expectation-maximization algorithm for estimating the parameter of a Markov modulated Poisson process are addressed. First, a scaling procedure is developed for the forward-backward recursions that circumvents the need for customized floating-point software. Second, evaluation of integrals of matrix exponentials is facilitated by applying a result due to Van Loan. For an MMPP of order four, a speedup of over two orders of magnitude was observed.

Linear predictive hidden Markov modeling is compared with a simple form of the switching autoregressive process. The latter process captures existing signal correlation during transitions of the Markov chain. Parameter estimation is described using naturally stable forward-backward recursions. The switching autoregressive model outperformed the linear predictive model in a digit recognition task and provided comparable performance to a cepstral-based recognizer.

We derive an explicit expression for the covariance of the log-periodogram power spectral density estimator for a zero mean Gaussian process. We do not make the assumption that the spectral components of the process are uncorrelated. Applications to spectral estimation and to cepstral modeling in automatic speech recognition are discussed.

Speaker classification is seen as a hypothesis testing problem of J simple hypotheses and a composite hypothesis. The simple hypotheses represent target speakers while the composite hypothesis represents nontarget speakers. The simple hypotheses have well-defined distributions that are estimated from training signals. The distribution of the signal under the composite hypothesis is assumed to belong to a given family. The parameter of that distribution is assumed random with a prior distribution that is estimated from a large set of speakers. This formulation converts the problem to that of testing J+1 simple hypotheses. Signals corresponding to target and nontarget speakers are assumed Gaussian mixtures processes. Once the system has been trained, list decoding is applied in which a test signal is associated with a list of possible speakers. The probability that the correct speaker is on the list is maximized for a given average number of incorrect speakers on the list. Results from speaker identification and speaker verification experiments are reported. In speaker identification using a National Institute of Standards and Technology (NIST) database with 174 target speakers, over 77% correct identification was achieved for an average of less than two erroneous speakers on the list. Speaker verification experiments on a similar database yielded results, expressed in terms of the equal-error-rate, of 6.7% and 10.1% using two decision rules.

We address the problem of estimating parameters of K-distribution. A recursive procedure based on the recursive EM algorithm is derived to find the ML estimates. Recursive EM is a stochastic approximation procedure with a gain matrix derived from the augmented data. Under mild conditions estimates generated by such procedure are characterized by strong consistency and asymptotic normality. Because of the simple structure of the augmented data, the proposed algorithm has a simple implementation. Numerical results show that the proposed approach performs well for various parameter sets.

In 1989 a new algorithm for non-iterative global maximization of the joint likelihood function of state and observation sequences of left-right HMM's was developed by Faragó and Lugosi. The algorithm capitalizes on the fact that the state sequence of a left-right HMM is uniquely determined by the state duration occupancies, The algorithm is mostly suitable for parameter estimation from a single training sequence. Extensions to estimation from multiple sequences are possible but deemed impractical. Two alternatives are proposed for utilizing this algorithm in automatic speech recognition. The first is based on averaging the parameter estimates from individual sequences while the second uses the Faragó and Lugosi segmentation to initialize the segmental k-means or the Baum algorithm. We have implemented the algorithm and tested it in isolated digit recognition. Using the second approach, a reduction of the error rate from .65% to .36% was realized

The optimal linear preprocessor for classifying two zero-mean Gaussian discrete-time signals which have been corrupted by additive zero-mean Gaussian noise is studied. Conditions for existence of the optimal linear preprocessor that achieves the performance of the likelihood ratio test for the noisy signals are given and the preprocessor is explicitly derived.

Robust speech recognition via hidden Markov modeling of spectral vectors is studied in this paper.The hidden Markov model (HMM) mixture components are assumed complex Gaussian with zero mean, diagonal covariance, and with incorporating an unknown scalar gain term.The gain term is associated with each spectral vector and it models the varying energy of speech signals.It is estimated by applying the maximum likelihood (ML) criterion.On an isolated digit database, in clean conditions, the spectral modeling with ML gain estimation approach a c hieved similar performance to cepstral modeling of speech.Two additive noise compensation approaches for the spectral modeling scheme are also considered.The rst approach requires a full noise HMM.This HMM is combined with the clean speech HMM to yield a noisy speech HMM.The second approach requires only the spectral shape of the noise.A term dependent on the spectral shape, together with an unknown magnitude term, is incorporated into the clean speech HMM to yield a noisy speech HMM.The unknown magnitude of the noise is estimated via the ML criterion.The performance of these two approaches for isolated digit recognition in noise is demonstrated and compared to a robust cepstral modeling approach from the literature.

The problem of identifying a speaker from a given utterance has been conventionally addressed using techniques such as Gaussian mixture models (GMMs) that model the characteristics of a known speaker via means and covariances. We compare the performance of a genetically optimised neural network speaker identification system versus the conventional approach of GMMs. The test data used in the experiments was the data used for the 1996 National Institute for Standards Technology (NIST) evaluation of speaker identification systems.