Multi-eigenspace normalization for robust speech recognition in noisy environments

Yoonjae Lee, Hanseok Ko

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

Abstract

In this paper, we propose an effective feature normalization scheme based on eigenspace normalization, for achieving robust speech recognition. In general, Mean and Variance Normalization (MVN) is implemented in cepstral domain. However, another MVN approach using eigenspace was recently introduced, in that the eigenspace normalization procedure performs normalization in a single eigenspace. This procedure consists of linear PCA matrix feature transformation followed by mean and variance normalization of the transformed cepstral feature. In the proposed scheme, we apply independent and unique eigenspaces to cepstra, delta and delta-delta cepstra respectively. We also normalize training data in eigenspace. In addition, a feature space rotation procedure is introduced to reduce the mismatch of training and test data distribution in noisy condition. As a result, we obtained a substantial improvement over the basic eigenspace normalization.

Original languageEnglish
Title of host publication8th International Conference on Spoken Language Processing, ICSLP 2004
PublisherInternational Speech Communication Association
Pages2097-2100
Number of pages4
Publication statusPublished - 2004
Event8th International Conference on Spoken Language Processing, ICSLP 2004 - Jeju, Jeju Island, Korea, Republic of
Duration: 2004 Oct 42004 Oct 8

Other

Other8th International Conference on Spoken Language Processing, ICSLP 2004
CountryKorea, Republic of
CityJeju, Jeju Island
Period04/10/404/10/8

Fingerprint

normalization
Speech Recognition
Normalization
mismatch

ASJC Scopus subject areas

  • Language and Linguistics
  • Linguistics and Language

Cite this

Lee, Y., & Ko, H. (2004). Multi-eigenspace normalization for robust speech recognition in noisy environments. In 8th International Conference on Spoken Language Processing, ICSLP 2004 (pp. 2097-2100). International Speech Communication Association.

Multi-eigenspace normalization for robust speech recognition in noisy environments. / Lee, Yoonjae; Ko, Hanseok.

8th International Conference on Spoken Language Processing, ICSLP 2004. International Speech Communication Association, 2004. p. 2097-2100.

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

Lee, Y & Ko, H 2004, Multi-eigenspace normalization for robust speech recognition in noisy environments. in 8th International Conference on Spoken Language Processing, ICSLP 2004. International Speech Communication Association, pp. 2097-2100, 8th International Conference on Spoken Language Processing, ICSLP 2004, Jeju, Jeju Island, Korea, Republic of, 04/10/4.
Lee Y, Ko H. Multi-eigenspace normalization for robust speech recognition in noisy environments. In 8th International Conference on Spoken Language Processing, ICSLP 2004. International Speech Communication Association. 2004. p. 2097-2100
Lee, Yoonjae ; Ko, Hanseok. / Multi-eigenspace normalization for robust speech recognition in noisy environments. 8th International Conference on Spoken Language Processing, ICSLP 2004. International Speech Communication Association, 2004. pp. 2097-2100
@inproceedings{770f4d20497d4188bfdebf06942515ff,
title = "Multi-eigenspace normalization for robust speech recognition in noisy environments",
abstract = "In this paper, we propose an effective feature normalization scheme based on eigenspace normalization, for achieving robust speech recognition. In general, Mean and Variance Normalization (MVN) is implemented in cepstral domain. However, another MVN approach using eigenspace was recently introduced, in that the eigenspace normalization procedure performs normalization in a single eigenspace. This procedure consists of linear PCA matrix feature transformation followed by mean and variance normalization of the transformed cepstral feature. In the proposed scheme, we apply independent and unique eigenspaces to cepstra, delta and delta-delta cepstra respectively. We also normalize training data in eigenspace. In addition, a feature space rotation procedure is introduced to reduce the mismatch of training and test data distribution in noisy condition. As a result, we obtained a substantial improvement over the basic eigenspace normalization.",
author = "Yoonjae Lee and Hanseok Ko",
year = "2004",
language = "English",
pages = "2097--2100",
booktitle = "8th International Conference on Spoken Language Processing, ICSLP 2004",
publisher = "International Speech Communication Association",

}

TY - GEN

T1 - Multi-eigenspace normalization for robust speech recognition in noisy environments

AU - Lee, Yoonjae

AU - Ko, Hanseok

PY - 2004

Y1 - 2004

N2 - In this paper, we propose an effective feature normalization scheme based on eigenspace normalization, for achieving robust speech recognition. In general, Mean and Variance Normalization (MVN) is implemented in cepstral domain. However, another MVN approach using eigenspace was recently introduced, in that the eigenspace normalization procedure performs normalization in a single eigenspace. This procedure consists of linear PCA matrix feature transformation followed by mean and variance normalization of the transformed cepstral feature. In the proposed scheme, we apply independent and unique eigenspaces to cepstra, delta and delta-delta cepstra respectively. We also normalize training data in eigenspace. In addition, a feature space rotation procedure is introduced to reduce the mismatch of training and test data distribution in noisy condition. As a result, we obtained a substantial improvement over the basic eigenspace normalization.

AB - In this paper, we propose an effective feature normalization scheme based on eigenspace normalization, for achieving robust speech recognition. In general, Mean and Variance Normalization (MVN) is implemented in cepstral domain. However, another MVN approach using eigenspace was recently introduced, in that the eigenspace normalization procedure performs normalization in a single eigenspace. This procedure consists of linear PCA matrix feature transformation followed by mean and variance normalization of the transformed cepstral feature. In the proposed scheme, we apply independent and unique eigenspaces to cepstra, delta and delta-delta cepstra respectively. We also normalize training data in eigenspace. In addition, a feature space rotation procedure is introduced to reduce the mismatch of training and test data distribution in noisy condition. As a result, we obtained a substantial improvement over the basic eigenspace normalization.

UR - http://www.scopus.com/inward/record.url?scp=85009084402&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85009084402&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85009084402

SP - 2097

EP - 2100

BT - 8th International Conference on Spoken Language Processing, ICSLP 2004

PB - International Speech Communication Association

ER -