A Rapid Match Algorithm for Continuous Speech Recognition

A Rapid Match Algorithm for Continuous Speech Recognition Laurence S. Gillick and Robert Roth Dragon Systems, Inc. 90 Bridge St. Newton MA. 02158 Abstract This paper describes an algorithm for performing rapid match on continuous speech that makes it possible to rec- ognize sentences from an 842 word vocabulary on a desk- top 33 megahertz 80486 computer in near real time. This algorithm relies on a combination of smoothing and linear segmentation together with the notion of word start groups. It appears that the total computation required grows more slowly than linearly with the vocabulary size, so that larger vocabularies appear feasible, with only moderately enhanced hardware. The rapid match algorithm described here is closely related to the one that is used in DragonDictate, Dragon's commercial 30,000 word discrete utterance recog- nizer. rapid match module to obtain a short list of plausible exten- sions. The key ideas that the algorithm relies on are linear segmentation, smoothing, acoustic clustering, and word start groupings. In subsequent sections we shall elaborate on these ideas and explain their role in rapid match. We shall then report on some empirical results, having to do with a particular task that Dragon has chosen to use for develop- ment purposes: the dictation of mammography reports, us- ing a vocabulary of 842 words. Other rapid match algorithms that are quite different in character have also been described in the literature [2], [3], and [4]. 1. Introduction For it to be feasible to perform large vocabulary con- tinuous speech recognition in near real time on currently available, moderately priced hardware, computational com- promises appear to be essential. One obvious compromise is to avoid detailed consideration of certain hypotheses that "cursory" inspection would reveal to be exceedingly un- likely. But what constitutes cursory inspection? To put it somewhat differently, how can we perform a very quick computation that would allow us to throw out many of the hypotheses that are "obviously" false? In particular, need dynamic programming play a role in such a "rapid match" algorithm? In this paper we shall describe a strategy for performing this kind of computation in the context of continuous speech recognition. The algorithm is an extension of the rapid match procedure that is used in DragonDictate, Dragon's commercially available 30,000 word discrete utter- ance recognizer. In that system, the interface between the rapid match module and the recognizer is very straightfor- ward. Each time the user says something, the rapid match algorithm provides the recognizer with a relatively short list of words that it thinks might have been said -- typically be- tween 100 and 200 words -- and in fact, this list is usually supplied to the recognizer before the speaker has finished saying the word. The nature of the interface between the rapid match mod- ule and the recognizer is different when continuous speech is involved, because the recognizer must contemplate hypothe- ses that represent sequences of words, in the course of rec- ognizing an utterance. In the system we describe [1], …

Content truncated to 500 words