A Robust Grammar Focused Parser for Spontaneously Spoken Language Thesis Summary

The analysis of spoken language is widely considered to be a more challenging task than the analysis of written text. All of the difficulties of written language can generally be found in spoken language as well. Parsing spontaneous speech must, however, also deal with problems such as speech disfluencies, the looser notion of grammaticality, and the lack of clearly marked sentence boundaries. The contamination of the input with errors of a speech recognizer can further exacerbate these problems. Most natural language parsing algorithms are designed to analyze “clean” grammatical input. Because they reject any input which is found to be ungrammatical in even the slightest way, such parsers are unsuitable for parsing spontaneous speech, where completely grammatical input is the exception more than the rule. This thesis describes GLR*, a parsing system based on Tomita’s Generalized LR parsing algorithm, that was designed to be robust to two particular types of extra-grammaticality: noise in the input, and limited grammar coverage. GLR* attempts to overcome these forms of extra-grammaticality by ignoring the unparsable words and fragments and conducting a search for the maximal subset of the original input that is covered by the grammar. The parser is coupled with a beam search heuristic, that limits the combinations of skipped words considered by the parser, and ensures that the parser will operate within feasible time and space bounds. The developed parsing system includes several tools designed to address the difficulties of parsing spontaneous speech. To cope with high levels of ambiguity, we developed a statistical disambiguation module, in which probabilities are attached directly to the actions in the LR parsing table. The parser must also determine the “best” parse from among the different parsable subsets of an input. We thus designed a general framework for combining a collection of parse evaluation measures into an integrated heuristic for evaluating and ranking the parses produced by the GLR* parser. This framework was applied to a set of four parse scoring measures developed for the JANUS scheduling domain and the ATIS domain. We added a parse quality heuristic, that allows the parser to self-judge the quality of the parse chosen as best, and to detect cases in which important information is likely to have been skipped. To demonstrate its suitability to parsing spontaneous speech, the GLR* parser was integrated into the JANUS speech translation system. Our evaluations on both transcribed and speech recognized input have indicated that the version of the system that uses GLR* produces between 15% and 30% more acceptable translations, than a corresponding version that uses the original non-robust GLR parser. We also developed a version of GLR* that is suitable to parsing word lattices produced by the speech recognizer, and investigated how lattice parsing can potentially overcome errors of the speech recognizer and further improve end-to-end performance of the speech translation system.