Lexical Access With a Statistically-Derived Phonetic Network

LEXICAL ACCESS WITH A STATISTICALLY-DERIVED PHONETIC NETWORK Michael D. Riley and Andrej Ljolje AT&T Bell Laboratories Murray Hill, NJ 07974 ABSTRACT A probabilistic approach to lexieal access from a recognized phone sequence is presented. Lexical access is seen as finding the word se- quence that maximizes the lexical likelihood of a sequence of phones and durations as recognized by a phone recognizer. This is theoretically correct for minimum error rate recognition within the model presented and is intuitively pleasing since it means that the "confusion matrix" of the phone recognizer will be learned and its regularities exploited. The lexical likelihoods are estimated from training data provided by the phone recognizer using statistical decision trees. Classification trees are used to estimate the phone realiziation distributions and regression trees are used to estimate the phone duration distributions, We find they can capture effectively allophonic variation, alternative pronun- ciation, word co-articulation and segmental durations. We describe a simpified, but efficient implementation of these models to lexical access in the DARPA resource management recognitiion task. 1. INTRODUCTION We describe a new approach to lexical access in a phone- based speech recognition system. By "lexical access" we mean taking a sequence (or, more generally, a lattice) of phones and durations that is output by a phone recognizer and mapping it onto a word sequence (or, more generally, a lattice). In conventional word-based speech recognizers, segmental durations, word co-articulation and alternative pronunciations are usually poorly modelled if at all since the architecture is not convenient or efficient for exploiting these constraints. Phone-based recognition offers an attractive alternative from this point of view. Our approach will be to create a probabilistic model that provides the likelihood that a particular word se- quence gives rise to a particular phone sequence. This model will take into account allophonic variation, alternative pronunciation, word co-articulation and segmental durations. We then combine these lexical likelihoods with the acoustic likelihoods generated by the phone recognizer and priors from our language model to get an overall recognition model whose error rate we seek to minimize. We have taken this stochastic approach for two reasons. First, it provides a principled way to combine seemingly disparate information: (a) acoustic likelihoods, (b) segmental durations, (c) alternative pronunciations, and (d) the language model. Sec- ond, the availability of large speech corpora now allow the sta- tistical estimation of these probabilities. 2. PROBABILISTIC MODEL We form the probabilistic model as follows. Let w be a sequence of words, let y be a sequence of phones, let d be a sequence of durations, and let s be a (fixed) speech signal. Then P(wis) ~x ZP(sly,d) P(y,d[w) P(w). (2.1) y,d The lefthand side of this relation is the probability that a given speech signal corresponds to a particular word sequence. The word sequence that maximizes this term gives the minimum sentence error rate. The first factor on the righthand side gives the acoustic likelihoods provided by the phone recognizer. The second factor gives the lexical likelihoods to be provided by the …

Content truncated to 500 words