Statistical feature selection for isolated word recognition

A procedure for feature selection in isolated word recognition is discussed. The feature selection is performed in two steps. The first step takes into account the temporal correlation among feature vectors in order to obtain a transformation matrix which projects the initial template of N feature vectors to a new space where they are uncorrelated. This step gives a new template of M feature vectors, where M≪N. The second step takes into account the frequency discrimination features which discriminate each word of the vocabulary from the others or a set of them. An important characteristic of this process is that the new templates do not need time alignment with the references in the comparison step, avoiding the use of the dynamic time-warping process. The speech recognition results show a significant improvement in the recognition performance with a digit database and the confusable E-set