HMM based fast keyworld spotting algorithm with no garbage models

The problem of discriminating keyword and non-keyword speech which is important in wordspotting applications is addressed. We have shown that garbage models cannot reduce both the rejection and false alarm rates simultaneously. To achieve this we have proposed a new scoring and search method for HMM based wordspotting without garbage models. This is a simple forward search method which incorporates the duration modelling of the keyword for efficient discrimination of keyword and non-keyword speech. This method is computationally fast, which makes it suitable for real-time implementation. The results are reported on a speaker independent database containing 10 keywords embedded in 150 carrier sentences.

[1]  Chin-Hui Lee,et al.  Automatic recognition of keywords in unconstrained speech using hidden Markov models , 1990, IEEE Trans. Acoust. Speech Signal Process..

[2]  Chin-Hui Lee,et al.  A frame-synchronous network search algorithm for connected word recognition , 1989, IEEE Trans. Acoust. Speech Signal Process..

[3]  R. C. Rose,et al.  Keyword detection in conversational speech utterances using hidden Markov model based continuous speech recognition , 1995, Comput. Speech Lang..

[4]  Hervé Bourlard,et al.  Optimizing recognition and rejection performance in wordspotting systems , 1994, Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing.