Pattern recognition system.

摘要

Multiple similarities between phoneme feature data extracted from a word input signal and reference pattern data read out from a reference pattern memory (3) are calculated by a similarity calculation section (2). The multiple similarities are converted by a conversion filter (4) into posterior probabilities using parameter sets which are read out from a filter parameter bank (5) in accordance with the obtained multiple similarities. The posterior probabilities are subjected to matching with data stored in word dictionary (7) by a word matching section (6), thus discriminating the input word.