Exploiting correlations among models with application to large vocabulary speech recognition

In a typical speech recognition system, computing the match between an incoming acoustic string and many competing models is computationally expensive. Once the highest ranking models are identified, all other match scores are discarded. We propose to make use of all computed scores by means of statistical inference. We view the match between an incoming acoustic string s and a model M, as a random variable Y-t. The class-conditional distributions of (Y\,..., YN) can be studied offline by sampling, and then used in a variety of ways. For example, the means of these distributions give rise to a natural measure of distance between models. One of the most useful applications of these distributions is as a basis for a new Bayesian classifier. The latter can be used to significantly reduce search effort in large vocabularies, and to quickly obtain a short list of candidate words. An example HMM-based system shows promising results. This research was supported by the Defense Advanced Research Projects Agency and monitored by the Space and Naval Warfare Systems Command under Contract N00039-9l-C-0158, ARPA Order No. 7239. The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of DARPA or the U.S. government.