Fourth European Conference on Speech Communication and Technology

Madrid, Spain
September 18-21, 1995

A Vocabulary Independent Discriminatively Trained Method for Rejection of Non-Keywords in Sub Word Based Speech Recognition

Rafid A. Sukkar, Chin-Hui Lee, Bling-Hwang Juang

AT&T Bell Laboratories, Naperville, IL, and Murray Hill, NJ, USA

Two important features of any deployable speech recognition system are the capability to detect if the input speech does not contain any of the recognizer keywords, and spot a keyword embedded in other speech or extraneous sounds. As a result, utterance verification (or non-keyword rejection) is becoming increasingly important as speech recognition systems continue to migrate from the laboratory to actual applications. In this paper we present a framework and a method for vocabulary independent utterance verification in subword based speech recognition. The verification process is cast as a statistical hypothesis test, where vocabulary independence is accomplished through a two stage verification process: subword level verification followed by string level verification. Experimental results show that this vocabulary independent discriminative utterance verification method significantly outperforms a baseline method commonly used in wordspotting tasks.

Full Paper

Bibliographic reference.  Sukkar, Rafid A. / Lee, Chin-Hui / Juang, Bling-Hwang (1995): "A vocabulary independent discriminatively trained method for rejection of non-keywords in sub word based speech recognition", In EUROSPEECH-1995, 1629-1632.