Modelling Context Dependency in Acoustic-Phonetic and Lexical Representations

Modelling Context Dependency in Acoustic-Phonetic and Lexical Representations 1 Michael Phillips, James Glass, and Victor Zue Spoken Language Systems Group Laboratory for Computer Science Massachusetts Institute of Technology Cambridge, Massachusetts 02139 ABSTRACT In 1989, our group first reported on the development of SUM- MIT, a segment-based speaker-independent continuous-speech re- cognition system [13] . The initial version of SUMMIT made use of fairly simple context-independent models for the lexical labels. Recently, we have begun to incorporate more complex models of lexical labels that take into account a variety of contextual fac- tors. These changes, along with an improved corrective training procedure for adapting pronunciation arc weights and a larger set of training data, have resulted in the reduction of error rate by almost a factor of two on the Resource Management task. INTRODUCTION Variability in speech arises from many different sources. For example, acoustic variability can be due to noise or chain nel characteristics, phonetic variability can be due to contex- tual or speaker-specific effects, and dialect effects can alter speakers' pronunciations of words. Speech recognition sys- tems must have mechanisms to model these various types of variability, and sometimes it may be necessary to deal with different types of variability with different mechanisms. For example, it may be difficult to find a single model that is able to deal effectively with both low-level acoustic variability and dialect differences among speakers. find mechanisms that are able to account for many different types of contextual factors. In this paper, we will describe a number of experiments in- tended to address some of the problems mentioned above. So far, we have attempted to account for some of the contextual effects on our phonetic models, although the approach that we have taken should apply to the higher levels of the system also. Briefly, we have found that we can increase recogni- tion performance by creating context-specific models or by using more flexible models. However, we did not see a per- formance increase when we combined the two in a straight- forward manner, presumably due to the fact that more flex- ible models tend to require more training data. If, instead of using context-specific models, we accounted for context by adjusting the input to the phonetic models (creating a context-normalized input vector), we were able to account for contextual effects and were able to use more flexible pho- netic models, resulting in the highest performance for our system. In the following sections, we will first provide an overview of the system. This will be followed by a more detailed de- scription of the changes we have made to the system, and evaluation results on the Resource Management task. In the SUMMIT system, we have made a rough distinc- tion between the sort of variability that we can deal with within our phonetic models (including acoustic variability and speaker differences at a phonetic level), and higher level phonological variation (including dialect effects and word- boundary effects). In both cases, our goal is to account for as …

Content truncated to 500 words