IEICE TRANSACTIONS on Information
Linear Discriminant Analysis Using a Generalized Mean of Class Covariances and Its Application to Speech Recognition
Summary :
To precisely model the time dependency of features is one of the important issues for speech recognition. Segmental unit input HMM with a dimensionality reduction method has been widely used to address this issue. Linear discriminant analysis (LDA) and heteroscedastic extensions, e.g., heteroscedastic linear discriminant analysis (HLDA) or heteroscedastic discriminant analysis (HDA), are popular approaches to reduce dimensionality. However, it is difficult to find one particular criterion suitable for any kind of data set in carrying out dimensionality reduction while preserving discriminative information. In this paper, we propose a new framework which we call power linear discriminant analysis (PLDA). PLDA can be used to describe various criteria including LDA, HLDA, and HDA with one control parameter. In addition, we provide an efficient selection method using a control parameter without training HMMs nor testing recognition performance on a development data set. Experimental results show that the PLDA is more effective than conventional methods for various data sets.
- Publication
- IEICE TRANSACTIONS on Information Vol.E91-D No.3 pp.478-487
- Publication Date
- 2008/03/01
- Publicized
- Online ISSN
- 1745-1361
- DOI
- 10.1093/ietisy/e91-d.3.478
- Type of Manuscript
- Special Section PAPER (Special Section on Robust Speech Processing in Realistic Environments)
- Category
- Feature Extraction
Authors
Keyword
Latest Issue
Copyrights notice of machine-translated contents
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Cite this
Copy
Makoto SAKAI, Norihide KITAOKA, Seiichi NAKAGAWA, "Linear Discriminant Analysis Using a Generalized Mean of Class Covariances and Its Application to Speech Recognition" in IEICE TRANSACTIONS on Information,
vol. E91-D, no. 3, pp. 478-487, March 2008, doi: 10.1093/ietisy/e91-d.3.478.
Abstract: To precisely model the time dependency of features is one of the important issues for speech recognition. Segmental unit input HMM with a dimensionality reduction method has been widely used to address this issue. Linear discriminant analysis (LDA) and heteroscedastic extensions, e.g., heteroscedastic linear discriminant analysis (HLDA) or heteroscedastic discriminant analysis (HDA), are popular approaches to reduce dimensionality. However, it is difficult to find one particular criterion suitable for any kind of data set in carrying out dimensionality reduction while preserving discriminative information. In this paper, we propose a new framework which we call power linear discriminant analysis (PLDA). PLDA can be used to describe various criteria including LDA, HLDA, and HDA with one control parameter. In addition, we provide an efficient selection method using a control parameter without training HMMs nor testing recognition performance on a development data set. Experimental results show that the PLDA is more effective than conventional methods for various data sets.
URL: https://global.ieice.org/en_transactions/information/10.1093/ietisy/e91-d.3.478/_p
Copy
@ARTICLE{e91-d_3_478,
author={Makoto SAKAI, Norihide KITAOKA, Seiichi NAKAGAWA, },
journal={IEICE TRANSACTIONS on Information},
title={Linear Discriminant Analysis Using a Generalized Mean of Class Covariances and Its Application to Speech Recognition},
year={2008},
volume={E91-D},
number={3},
pages={478-487},
abstract={To precisely model the time dependency of features is one of the important issues for speech recognition. Segmental unit input HMM with a dimensionality reduction method has been widely used to address this issue. Linear discriminant analysis (LDA) and heteroscedastic extensions, e.g., heteroscedastic linear discriminant analysis (HLDA) or heteroscedastic discriminant analysis (HDA), are popular approaches to reduce dimensionality. However, it is difficult to find one particular criterion suitable for any kind of data set in carrying out dimensionality reduction while preserving discriminative information. In this paper, we propose a new framework which we call power linear discriminant analysis (PLDA). PLDA can be used to describe various criteria including LDA, HLDA, and HDA with one control parameter. In addition, we provide an efficient selection method using a control parameter without training HMMs nor testing recognition performance on a development data set. Experimental results show that the PLDA is more effective than conventional methods for various data sets.},
keywords={},
doi={10.1093/ietisy/e91-d.3.478},
ISSN={1745-1361},
month={March},}
Copy
TY - JOUR
TI - Linear Discriminant Analysis Using a Generalized Mean of Class Covariances and Its Application to Speech Recognition
T2 - IEICE TRANSACTIONS on Information
SP - 478
EP - 487
AU - Makoto SAKAI
AU - Norihide KITAOKA
AU - Seiichi NAKAGAWA
PY - 2008
DO - 10.1093/ietisy/e91-d.3.478
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E91-D
IS - 3
JA - IEICE TRANSACTIONS on Information
Y1 - March 2008
AB - To precisely model the time dependency of features is one of the important issues for speech recognition. Segmental unit input HMM with a dimensionality reduction method has been widely used to address this issue. Linear discriminant analysis (LDA) and heteroscedastic extensions, e.g., heteroscedastic linear discriminant analysis (HLDA) or heteroscedastic discriminant analysis (HDA), are popular approaches to reduce dimensionality. However, it is difficult to find one particular criterion suitable for any kind of data set in carrying out dimensionality reduction while preserving discriminative information. In this paper, we propose a new framework which we call power linear discriminant analysis (PLDA). PLDA can be used to describe various criteria including LDA, HLDA, and HDA with one control parameter. In addition, we provide an efficient selection method using a control parameter without training HMMs nor testing recognition performance on a development data set. Experimental results show that the PLDA is more effective than conventional methods for various data sets.
ER -
English
