A Bayesian-adaptive decision method for the V/UV/S classification of segments of a speech signal

In this correspondence, a method for voiced (V), unvoiced (UV), or silence (S) classification of speech segments, based on the maximum a posteriori probability criterion, is presented. The a poste-riori probabilities of the three classes are determined using a vector x = (fi,. .. ,ff) of measurements on the segment under consideration. It is assumed that the vector x has an L-dimensional Gaussian distribution with an expected random value also characterized by an L-dimensional Gaussian distribution. In addition, it is assumed that the sequence of the classes constitutes a first-order stationary Markov chain. The initial parameters are estimated in a training phase. During the application phase, the decision method is adapted by using the previous classifications in order to update the probability density function (pdf) of the expected random values. I. INTRODUCTION In the past, various methods have been proposed for the automatic V/UV/S classification: some are based on a probabilistic approach , while others are of the deterministic type. Among the prob-abilistic methods, we recall in particular the one proposed by Atal and Rabiner [I] in which a minimum distance classifier, based on the maximum likelihood criterion, is used. The method proposed in this correspondence is probabilistic and is based on a Bayesian approach. The maximum a posteriori probability criterion is used. In addition, the method presented is adap-tive: some pdf's, estimated in a training phase, are updated taking into account the previous decision. At each step, the classification is based on a vector of measure