Latent semantic information in maximum entropy language models for conversational speech recognition

Authors:
Yonggang Deng;Sanjeev Khudanpur
Affiliations:
The Johns Hopkins University, Baltimore, MD;The Johns Hopkins University, Baltimore, MD
Venue:
NAACL '03 Proceedings of the 2003 Conference of the North American Chapter of the Association for Computational Linguistics on Human Language Technology - Volume 1
Year:
2003

Citing 4
Cited 1

Language Model Adaptation Using Mixtures and an Exponentially Decaying Cache

ICASSP '97 Proceedings of the 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP '97)-Volume 2 - Volume 2
SVDPACKC (Version 1.0) User''s Guide

SVDPACKC (Version 1.0) User''s Guide
Maximum entropy language modeling with non-local dependencies

Maximum entropy language modeling with non-local dependencies
A maximum entropy language model integrating N-grams and topic dependencies for conversational speech recognition

ICASSP '99 Proceedings of the Acoustics, Speech, and Signal Processing, 1999. on 1999 IEEE International Conference - Volume 01

Language models based on semantic composition

EMNLP '09 Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing: Volume 1 - Volume 1

Quantified Score

Hi-index	0.00

Visualization

Abstract

Latent semantic analysis (LSA), first exploited in indexing documents for information retrieval, has since been used by several researchers to demonstrate impressive reductions in the perplexity of statistical language models on text corpora such as the Wall Street Journal. In this paper we present an investigation into the use of LSA in language modeling for conversational speech recognition. We find that previously proposed methods of combining an LSA-based unigram model with an N-gram model yield much smaller reductions in perplexity on speech transcriptions than has been reported on written text. We next present a family of exponential models in which LSA similarity is a feature of a word-history pair. The maximum entropy model in this family yields a greater reduction in perplexity, and statistically significant improvements in recognition accuracy over a trigram model on the Switchboard corpus. We conclude with a comparison of this LSA-featured model with a previously proposed topic-dependent maximum entropy model.