Apparatus, method and system for cross-speaker speech recognition for telecommunication applications

Abstract

The apparatus, method and system of the present invention provide for cross-speaker speech recognition, and are particularly suited for telecommunication applications such as automatic name (voice) dialing, message management, call return management, and incoming call screening. The method of the present invention includes receiving incoming speech, such as an incoming caller name, and generating a phonetic transcription of the incoming speech with a speaker-independent, hidden Markov model having an unconstrained grammar in which any phoneme may follow any other phoneme, followed by determining a transcription parameter as a likelihood of fit of the incoming speech to the speaker-independent model. The method further selects a first phoneme pattern, from a plurality of phoneme patterns, as having a highest likelihood of fit to the incoming speech, utilizing a speaker-independent, hidden Markov model having a grammar constrained by these phoneme patterns, followed by determining a recognition parameter as a likelihood of fit of the incoming speech to the selected, first phoneme pattern. The method then determines whether the input speech matches or collides with the first phoneme pattern based upon a correspondence of the transcription parameter with the recognition parameter in accordance with a predetermined criterion. In the preferred embodiment, this matching or collision determination is made as a function of a confidence ratio, the ratio of the transcription parameter to the recognition parameter, being within or less than a predetermined threshold value.