Spatial histogram equalization of complex-valued acoustic spectra in modulation domain for noise-robust speech recognition

Hsin Ju Hsieh, Berlin Chen, Jeih Weih Hung

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

1 Citation (Scopus)

Abstract

This paper proposes to enhance the complex-valued acoustic spectrograms of speech signals via the technique of histogram equalization (HEQ) to produce noise-robust features for recognition. The presented method extends our previous work in the task of spectrogram enhancement and has two significant aspects. First, we process the real and imaginary parts of acoustic spectrograms separately, and therefore both of the corresponding magnitude and phase components can be enhanced implicitly. Second, we apply FIR filters to the intra-frame acoustic spectra to acquire the respective local structural statistics, which are subsequently employed to perform various types of HEQ on the acoustic spectrograms for robustifying the resulting speech features. All experiments were carried out on the Aurora-2 database and task. The performance of the presented methods was thoroughly tested and verified by comparisons with other well-known robustness methods, which reveals the capability of our methods in promoting the noise robustness of speech features.

Original languageEnglish
Title of host publication2014 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2014
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9786163618238
DOIs
Publication statusPublished - 2014 Feb 12
Event2014 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2014 - Chiang Mai, Thailand
Duration: 2014 Dec 92014 Dec 12

Publication series

Name2014 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2014

Other

Other2014 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2014
CountryThailand
CityChiang Mai
Period14/12/914/12/12

Fingerprint

Speech recognition
Acoustic noise
Acoustics
Modulation
FIR filters
Statistics
Experiments

ASJC Scopus subject areas

  • Signal Processing
  • Information Systems

Cite this

Hsieh, H. J., Chen, B., & Hung, J. W. (2014). Spatial histogram equalization of complex-valued acoustic spectra in modulation domain for noise-robust speech recognition. In 2014 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2014 [7041568] (2014 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2014). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/APSIPA.2014.7041568

Spatial histogram equalization of complex-valued acoustic spectra in modulation domain for noise-robust speech recognition. / Hsieh, Hsin Ju; Chen, Berlin; Hung, Jeih Weih.

2014 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2014. Institute of Electrical and Electronics Engineers Inc., 2014. 7041568 (2014 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2014).

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

Hsieh, HJ, Chen, B & Hung, JW 2014, Spatial histogram equalization of complex-valued acoustic spectra in modulation domain for noise-robust speech recognition. in 2014 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2014., 7041568, 2014 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2014, Institute of Electrical and Electronics Engineers Inc., 2014 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2014, Chiang Mai, Thailand, 14/12/9. https://doi.org/10.1109/APSIPA.2014.7041568
Hsieh HJ, Chen B, Hung JW. Spatial histogram equalization of complex-valued acoustic spectra in modulation domain for noise-robust speech recognition. In 2014 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2014. Institute of Electrical and Electronics Engineers Inc. 2014. 7041568. (2014 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2014). https://doi.org/10.1109/APSIPA.2014.7041568
Hsieh, Hsin Ju ; Chen, Berlin ; Hung, Jeih Weih. / Spatial histogram equalization of complex-valued acoustic spectra in modulation domain for noise-robust speech recognition. 2014 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2014. Institute of Electrical and Electronics Engineers Inc., 2014. (2014 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2014).
@inproceedings{639c58c27fc54e82b741c0fe66fef675,
title = "Spatial histogram equalization of complex-valued acoustic spectra in modulation domain for noise-robust speech recognition",
abstract = "This paper proposes to enhance the complex-valued acoustic spectrograms of speech signals via the technique of histogram equalization (HEQ) to produce noise-robust features for recognition. The presented method extends our previous work in the task of spectrogram enhancement and has two significant aspects. First, we process the real and imaginary parts of acoustic spectrograms separately, and therefore both of the corresponding magnitude and phase components can be enhanced implicitly. Second, we apply FIR filters to the intra-frame acoustic spectra to acquire the respective local structural statistics, which are subsequently employed to perform various types of HEQ on the acoustic spectrograms for robustifying the resulting speech features. All experiments were carried out on the Aurora-2 database and task. The performance of the presented methods was thoroughly tested and verified by comparisons with other well-known robustness methods, which reveals the capability of our methods in promoting the noise robustness of speech features.",
author = "Hsieh, {Hsin Ju} and Berlin Chen and Hung, {Jeih Weih}",
year = "2014",
month = "2",
day = "12",
doi = "10.1109/APSIPA.2014.7041568",
language = "English",
series = "2014 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2014",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
booktitle = "2014 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2014",

}

TY - GEN

T1 - Spatial histogram equalization of complex-valued acoustic spectra in modulation domain for noise-robust speech recognition

AU - Hsieh, Hsin Ju

AU - Chen, Berlin

AU - Hung, Jeih Weih

PY - 2014/2/12

Y1 - 2014/2/12

N2 - This paper proposes to enhance the complex-valued acoustic spectrograms of speech signals via the technique of histogram equalization (HEQ) to produce noise-robust features for recognition. The presented method extends our previous work in the task of spectrogram enhancement and has two significant aspects. First, we process the real and imaginary parts of acoustic spectrograms separately, and therefore both of the corresponding magnitude and phase components can be enhanced implicitly. Second, we apply FIR filters to the intra-frame acoustic spectra to acquire the respective local structural statistics, which are subsequently employed to perform various types of HEQ on the acoustic spectrograms for robustifying the resulting speech features. All experiments were carried out on the Aurora-2 database and task. The performance of the presented methods was thoroughly tested and verified by comparisons with other well-known robustness methods, which reveals the capability of our methods in promoting the noise robustness of speech features.

AB - This paper proposes to enhance the complex-valued acoustic spectrograms of speech signals via the technique of histogram equalization (HEQ) to produce noise-robust features for recognition. The presented method extends our previous work in the task of spectrogram enhancement and has two significant aspects. First, we process the real and imaginary parts of acoustic spectrograms separately, and therefore both of the corresponding magnitude and phase components can be enhanced implicitly. Second, we apply FIR filters to the intra-frame acoustic spectra to acquire the respective local structural statistics, which are subsequently employed to perform various types of HEQ on the acoustic spectrograms for robustifying the resulting speech features. All experiments were carried out on the Aurora-2 database and task. The performance of the presented methods was thoroughly tested and verified by comparisons with other well-known robustness methods, which reveals the capability of our methods in promoting the noise robustness of speech features.

UR - http://www.scopus.com/inward/record.url?scp=84949926199&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84949926199&partnerID=8YFLogxK

U2 - 10.1109/APSIPA.2014.7041568

DO - 10.1109/APSIPA.2014.7041568

M3 - Conference contribution

AN - SCOPUS:84949926199

T3 - 2014 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2014

BT - 2014 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2014

PB - Institute of Electrical and Electronics Engineers Inc.

ER -