About cookies on this site Our websites require some cookies to function properly (required). In addition, other cookies may be used with your consent to analyze site usage, improve the user experience and for advertising. For more information, please review your options. By visiting our website, you agree to our processing of information as described in IBM’sprivacy statement. To provide a smooth navigation, your cookie preferences will be shared across the IBM web domains listed here.
Paper
Optimal Solution of a Training Problem in Speech Recognition
Abstract
We take the view that the payoff correpsonding to different ways of training a speech recognizer is the probability that the recognizer will correctly recognize a randomly chosen word. In “A Decision Theoretic Formulation of a Training Problem in Speech Recognition”1 we posed the problem of choosing a speech recognizer as an optimization problem with the expected value of the above payoff as the objective function. This correspondence presents the optimal Bayes solution to this optimization problem by maximizing the expected payoff: conditionally on given training data decode the acoustic signal for a word as any word which maximizes the a posteriori expected joint probability of the word and the acoustic signal. Thus the probability estimator which is optimal for mean-squared error produces a decoder which happens to be optimal for recognition rate as well. Copyright © 1985 by The Institute of Electrical and Electronics Engineers, Inc.