In recent years, many alternative models have been proposed to address some of the shortcomings of the hidden Markov model, currently the most popular approach to speech recognition. In particular, a variety of models that could be broadly classi ed as segment models have been described for representing a variable-length sequence of observation vectors in speech recognition applications. Since there are many aspects in common between these approaches, including the general recognition and training problems, it is useful to consider them in a uni ed framework. Thus, the goal of this paper will be to describe a general stochastic model that encompasses most of the models proposed in the literature, pointing out similarities of the models in terms of correlation and parameter tying assumptions, and drawing analogies between segment models and hidden Markov models. In addition, we summarize experimental results assessing di erent modeling assumptions, and point out remaining open questions.
In this paper we present a nontraditional approach to the problem of estimating the parameters of a stochastic linear system. The method is based on the Expectation-Maximization algorithm and can be considered as the continuous analog of the Baum-Welch estimation algorithm for hidden Markov models. We use the algorithm for training the parameters of a dynamical system model that we propose for better representing the spectral dynamics of speech for recognition. We assume that the observed feature vectors of a phone segment are the output of a stochastic linear dynamical system, and we show how the evolution of the dynamics as a function of the segment length can be modeled using alternative assumptions. We show on a phoneme classification task using the TIMIT database that our approach is the first effective use of an explicit model for statistical dependence between frames of speech.
The outbreak of COVID‐19 led to a record‐breaking race to develop a vaccine. However, the limited vaccine capacity creates another massive challenge: how to distribute vaccines to mitigate the near‐end impact of the pandemic? In the United States in particular, the new Biden administration is launching mass vaccination sites across the country, raising the obvious question of where to locate these clinics to maximize the effectiveness of the vaccination campaign. This paper tackles this question with a novel data‐driven approach to optimize COVID‐19 vaccine distribution. We first augment a state‐of‐the‐art epidemiological model, called DELPHI, to capture the effects of vaccinations and the variability in mortality rates across age groups. We then integrate this predictive model into a prescriptive model to optimize the location of vaccination sites and subsequent vaccine allocation. The model is formulated as a bilinear, nonconvex optimization model. To solve it, we propose a coordinate descent algorithm that iterates between optimizing vaccine distribution and simulating the dynamics of the pandemic. As compared to benchmarks based on demographic and epidemiological information, the proposed optimization approach increases the effectiveness of the vaccination campaign by an estimated 20%, saving an extra 4000 extra lives in the United States over a 3‐month period. The proposed solution achieves critical fairness objectives—by reducing the death toll of the pandemic in several states without hurting others—and is highly robust to uncertainties and forecast errors—by achieving similar benefits under a vast range of perturbations.
Abstruci-An algorithm is proposed that achieves a good tradeoff between modeling resolution and robustness by using a new, general scheme for tying of mixture components in continuous mixture-density hidden Markov model (HMM)-based speech recognizers. The sets of HMM states that share the same mixture components are determined automatically using agglomerative clustering techniques. Experimental results on ARPA's Wall Street Journal corpus show that this scheme reduces errors by 25% over typical tied-mixture systems. New fast algorithms for computing Gaussian likelihoods-the most time-consuming aspect of continuous-density HMM systems-are also presented.These new algorithms significantly reduce the number of Gaussian densities that are evaluated with little or no impact on speech recognition accuracy.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.