Minimizing search errors due to delayed bigrams in real-time speech recognition systems

When building applications from large vocabulary speech recognition systems, a certain amount of search errors due to pruning often has to be accepted in order to obtain the required speed. In this paper we tackle the problems resulting from aggressive pruning strategies as typically applied in large vocabulary systems to achieve close to real-time performance. We consider a typical scenario of a two pass viterbi search with the rst pass being organized as a phoneme (allophone) tree. For such a tree organized lexicon, there are two possiblities to use a bigram language model: either by building tree copies or by using so-called delayed bigrams. Since copying trees turns out to be too expensive for real time applications we basically refer to delayed bigrams, discuss their drastic in uence on the word accuracy and show how to alleviate the desastrous e ect of delayed bigrams under aggressive pruning.