Hiroyuki Iwahashi

Voice recognition method and apparatus by updating reference patterns

Inputted voice signals are analyzed in units of syllables. Each syllable is compared with standard syllables preregistered in a memory and thereby recognized but corrections may be made on erroneous recognitions by referencing a dictionary or entering a command. Each standard pattern is associated with phonological information on the neighborhood in a voice signal from where it was extracted and may be updated by another pattern having the same phonological information. Temporal sequences of correct and erroneous recognitions of individual syllables as well as whole syllables are stored and referenced in determining whether a standard pattern should be updated. A maximum pattern number may be set for each syllable.

Phonemic classification in speech recognition system having accelerated response time

An acoustic processing unit for a speech recognition system is characterized in that the parameters are extracted from the speech data, including correlation functions, the zero crossing number of the original waveforms, the zero crossing number of the differential waveforms, and the average level of the waveforms. A suitable threshold is selected from a plurality of thresholds preliminarily stored, depending on the inputted speech volume level. The inputted speech volume level of the speaker is detected so that feedback of the detected volume level is obtained. The selected threshold set is then compared with each of the parameters to thereby make the phonemic classification. Because the plurality of threshold sets are thus automatically selected depending on the inputted speech volume level, flexible phonemic classification can be obtained to exactly detect the speech sections.