Frequency warping for speaker recognition

摘要

A voice recognition system comprises an analyzer 1 for converting an input voice signal to an input pattern including cepstrum, a reference pattern memory 3 for storing reference patterns, an elongation/contraction estimating unit 4 for outputting an elongation/contraction parameter in frequency axis direction by using the input pattern and the reference patterns, and a recognizing unit 101 for calculating the distances between the converted input pattern from the converter 2 and the reference patterns and outputting the reference pattern corresponding to the shortest distance as result of recognition. The elongation/contraction unit 4 estimates an elongation/contraction parameter by using cepstrum included in the input pattern. The unit 4 does not have various values in advance for determining the elongation/contraction parameter. Nor the unit 4 have to execute distance calculation for various values.