Generalized likelihood ratio discriminant analysis

Hung Shin Lee, Berlin Chen

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Citations (Scopus)

Abstract

In the past several decades, classifier-independent front-end feature extraction, where the derivation of acoustic features is lightly associated with the back-end model training or classification, has been prominently used in various pattern recognition tasks, including automatic speech recognition (ASR). In this paper, we present a novel discriminative feature transformation, named generalized likelihood ratio discriminant analysis (GLRDA), on the basis of the likelihood ratio test (LRT). It attempts to seek a lower dimensional feature subspace by making the most confusing situation, described by the null hypothesis, as unlikely to happen as possible without the homoscedastic assumption on class distributions. We also show that the classical linear discriminant analysis (LDA) and its well-known extension - heteroscedastic linear discriminant analysis (HLDA) can be regarded as two special cases of our proposed method. The empirical class confusion information can be further incorporated into GLRDA for better recognition performance. Experimental results demonstrate that GLRDA and its variant can yield moderate performance improvements over HLDA and LDA for the large vocabulary continuous speech recognition (LVCSR) task.

Original languageEnglish
Title of host publicationProceedings of the 2009 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2009
Pages158-163
Number of pages6
DOIs
Publication statusPublished - 2009 Dec 1
Event2009 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2009 - Merano, Italy
Duration: 2009 Dec 132009 Dec 17

Publication series

NameProceedings of the 2009 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2009

Other

Other2009 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2009
CountryItaly
CityMerano
Period09/12/1309/12/17

Fingerprint

Discriminant analysis
Continuous speech recognition
Speech recognition
Pattern recognition
Feature extraction
Classifiers
Acoustics

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Human-Computer Interaction
  • Signal Processing

Cite this

Lee, H. S., & Chen, B. (2009). Generalized likelihood ratio discriminant analysis. In Proceedings of the 2009 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2009 (pp. 158-163). [5373392] (Proceedings of the 2009 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2009). https://doi.org/10.1109/ASRU.2009.5373392

Generalized likelihood ratio discriminant analysis. / Lee, Hung Shin; Chen, Berlin.

Proceedings of the 2009 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2009. 2009. p. 158-163 5373392 (Proceedings of the 2009 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2009).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Lee, HS & Chen, B 2009, Generalized likelihood ratio discriminant analysis. in Proceedings of the 2009 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2009., 5373392, Proceedings of the 2009 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2009, pp. 158-163, 2009 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2009, Merano, Italy, 09/12/13. https://doi.org/10.1109/ASRU.2009.5373392
Lee HS, Chen B. Generalized likelihood ratio discriminant analysis. In Proceedings of the 2009 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2009. 2009. p. 158-163. 5373392. (Proceedings of the 2009 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2009). https://doi.org/10.1109/ASRU.2009.5373392
Lee, Hung Shin ; Chen, Berlin. / Generalized likelihood ratio discriminant analysis. Proceedings of the 2009 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2009. 2009. pp. 158-163 (Proceedings of the 2009 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2009).
@inproceedings{725569f3ac82410896009bb6457776f9,
title = "Generalized likelihood ratio discriminant analysis",
abstract = "In the past several decades, classifier-independent front-end feature extraction, where the derivation of acoustic features is lightly associated with the back-end model training or classification, has been prominently used in various pattern recognition tasks, including automatic speech recognition (ASR). In this paper, we present a novel discriminative feature transformation, named generalized likelihood ratio discriminant analysis (GLRDA), on the basis of the likelihood ratio test (LRT). It attempts to seek a lower dimensional feature subspace by making the most confusing situation, described by the null hypothesis, as unlikely to happen as possible without the homoscedastic assumption on class distributions. We also show that the classical linear discriminant analysis (LDA) and its well-known extension - heteroscedastic linear discriminant analysis (HLDA) can be regarded as two special cases of our proposed method. The empirical class confusion information can be further incorporated into GLRDA for better recognition performance. Experimental results demonstrate that GLRDA and its variant can yield moderate performance improvements over HLDA and LDA for the large vocabulary continuous speech recognition (LVCSR) task.",
author = "Lee, {Hung Shin} and Berlin Chen",
year = "2009",
month = "12",
day = "1",
doi = "10.1109/ASRU.2009.5373392",
language = "English",
isbn = "9781424454792",
series = "Proceedings of the 2009 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2009",
pages = "158--163",
booktitle = "Proceedings of the 2009 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2009",

}

TY - GEN

T1 - Generalized likelihood ratio discriminant analysis

AU - Lee, Hung Shin

AU - Chen, Berlin

PY - 2009/12/1

Y1 - 2009/12/1

N2 - In the past several decades, classifier-independent front-end feature extraction, where the derivation of acoustic features is lightly associated with the back-end model training or classification, has been prominently used in various pattern recognition tasks, including automatic speech recognition (ASR). In this paper, we present a novel discriminative feature transformation, named generalized likelihood ratio discriminant analysis (GLRDA), on the basis of the likelihood ratio test (LRT). It attempts to seek a lower dimensional feature subspace by making the most confusing situation, described by the null hypothesis, as unlikely to happen as possible without the homoscedastic assumption on class distributions. We also show that the classical linear discriminant analysis (LDA) and its well-known extension - heteroscedastic linear discriminant analysis (HLDA) can be regarded as two special cases of our proposed method. The empirical class confusion information can be further incorporated into GLRDA for better recognition performance. Experimental results demonstrate that GLRDA and its variant can yield moderate performance improvements over HLDA and LDA for the large vocabulary continuous speech recognition (LVCSR) task.

AB - In the past several decades, classifier-independent front-end feature extraction, where the derivation of acoustic features is lightly associated with the back-end model training or classification, has been prominently used in various pattern recognition tasks, including automatic speech recognition (ASR). In this paper, we present a novel discriminative feature transformation, named generalized likelihood ratio discriminant analysis (GLRDA), on the basis of the likelihood ratio test (LRT). It attempts to seek a lower dimensional feature subspace by making the most confusing situation, described by the null hypothesis, as unlikely to happen as possible without the homoscedastic assumption on class distributions. We also show that the classical linear discriminant analysis (LDA) and its well-known extension - heteroscedastic linear discriminant analysis (HLDA) can be regarded as two special cases of our proposed method. The empirical class confusion information can be further incorporated into GLRDA for better recognition performance. Experimental results demonstrate that GLRDA and its variant can yield moderate performance improvements over HLDA and LDA for the large vocabulary continuous speech recognition (LVCSR) task.

UR - http://www.scopus.com/inward/record.url?scp=77949417964&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77949417964&partnerID=8YFLogxK

U2 - 10.1109/ASRU.2009.5373392

DO - 10.1109/ASRU.2009.5373392

M3 - Conference contribution

AN - SCOPUS:77949417964

SN - 9781424454792

T3 - Proceedings of the 2009 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2009

SP - 158

EP - 163

BT - Proceedings of the 2009 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2009

ER -