In a system for recognizing a time sequence of feature vectors of a speech signal representative of an unknown utterance as one of a plurality of reference patterns, a generator (11) for generating the reference patterns has a converter (15) for converting a plurality of time sequences of feature vectors of an input pattern of a speech signal with variances to a plurality of time sequences of feature codes with reference to code vectors (14) which are previously prepared by the known clustering. A first pattern former (16) generates a state transition probability distribution and an occurrence probability distribution of feature codes for each state in a state transition network. A function generator (17) calculates parameters of continuous Gaussian density function from the code vectors and the occurrence probability distribution to produce the continuous Gaussian density function approximating the occurrence probability distribution. A second pattern former (18) produces a reference pattern defined by the state transition probability distribution and the continuous Gaussian density function. For a plurality of different training words, a plurality of reference patterns are generated and are memorized in the reference pattern generator.