Fast approximation of Karhunen-Loeve transform for problems of speech-like signal recognition

The problem of speech recognition becomes more and more important, and one of the difficult tasks in that problem is the transformation of an initial signal to set of parameters to be recognized. The optimum orthogonal transform which is the best preserved information is given by the augmented eigenvectors of the data domain covariance matrix. Such transformation, commonly called the Karhunen-Loeve transformation (KLT) is too complex to be installed into a speech processing system. By imposing some restrictions to the input signal, it is possible to simplify calculation of the KLT by its reduction to another transformation, which is easily calculated. Moreover, these restrictions allow one to use fast algorithms for calculation. Such a technique was applied to real speech signals and the correctness of the imposed restrictions was proved. The technique of speech signal transformation into a sequence of parameters has been developed and experimentally proved.