Stereo-based feature enhancement using dictionary learning

Shinji Watanabe, John R. Hershey

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

1 Citation (Scopus)

Abstract

This paper proposes stereo-based speech feature enhancement using dictionary learning. Instead of posterior values obtained by a Gaussian mixture as in other methods, we use sparse weight vectors and their variants as an alternative noisy speech feature representation. This paper also provides an efficient algorithm that can be applied to large-scale speech processing. We show the effectiveness of the proposed approach by using a middle vocabulary noisy speech recognition task based on WSJ, which was provided by the 2nd CHiME Speech Separation and Recognition Challenge.

Original languageEnglish
Title of host publication2013 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2013 - Proceedings
Pages7073-7077
Number of pages5
DOIs
Publication statusPublished - 2013 Oct 18
Externally publishedYes
Event2013 38th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2013 - Vancouver, BC
Duration: 2013 May 262013 May 31

Other

Other2013 38th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2013
CityVancouver, BC
Period13/5/2613/5/31

Fingerprint

Glossaries
Speech processing
Speech recognition

Keywords

  • 2nd CHiME challenge track 2
  • dictionary learning
  • sparse representation
  • speech feature enhancement
  • Speech recognition

ASJC Scopus subject areas

  • Signal Processing
  • Software
  • Electrical and Electronic Engineering

Cite this

Watanabe, S., & Hershey, J. R. (2013). Stereo-based feature enhancement using dictionary learning. In 2013 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2013 - Proceedings (pp. 7073-7077). [6639034] https://doi.org/10.1109/ICASSP.2013.6639034

Stereo-based feature enhancement using dictionary learning. / Watanabe, Shinji; Hershey, John R.

2013 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2013 - Proceedings. 2013. p. 7073-7077 6639034.

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

Watanabe, S & Hershey, JR 2013, Stereo-based feature enhancement using dictionary learning. in 2013 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2013 - Proceedings., 6639034, pp. 7073-7077, 2013 38th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2013, Vancouver, BC, 13/5/26. https://doi.org/10.1109/ICASSP.2013.6639034
Watanabe S, Hershey JR. Stereo-based feature enhancement using dictionary learning. In 2013 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2013 - Proceedings. 2013. p. 7073-7077. 6639034 https://doi.org/10.1109/ICASSP.2013.6639034
Watanabe, Shinji ; Hershey, John R. / Stereo-based feature enhancement using dictionary learning. 2013 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2013 - Proceedings. 2013. pp. 7073-7077
@inproceedings{93db49c254b04deba5d4134b376a9b53,
title = "Stereo-based feature enhancement using dictionary learning",
abstract = "This paper proposes stereo-based speech feature enhancement using dictionary learning. Instead of posterior values obtained by a Gaussian mixture as in other methods, we use sparse weight vectors and their variants as an alternative noisy speech feature representation. This paper also provides an efficient algorithm that can be applied to large-scale speech processing. We show the effectiveness of the proposed approach by using a middle vocabulary noisy speech recognition task based on WSJ, which was provided by the 2nd CHiME Speech Separation and Recognition Challenge.",
keywords = "2nd CHiME challenge track 2, dictionary learning, sparse representation, speech feature enhancement, Speech recognition",
author = "Shinji Watanabe and Hershey, {John R.}",
year = "2013",
month = "10",
day = "18",
doi = "10.1109/ICASSP.2013.6639034",
language = "English",
isbn = "9781479903566",
pages = "7073--7077",
booktitle = "2013 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2013 - Proceedings",

}

TY - GEN

T1 - Stereo-based feature enhancement using dictionary learning

AU - Watanabe, Shinji

AU - Hershey, John R.

PY - 2013/10/18

Y1 - 2013/10/18

N2 - This paper proposes stereo-based speech feature enhancement using dictionary learning. Instead of posterior values obtained by a Gaussian mixture as in other methods, we use sparse weight vectors and their variants as an alternative noisy speech feature representation. This paper also provides an efficient algorithm that can be applied to large-scale speech processing. We show the effectiveness of the proposed approach by using a middle vocabulary noisy speech recognition task based on WSJ, which was provided by the 2nd CHiME Speech Separation and Recognition Challenge.

AB - This paper proposes stereo-based speech feature enhancement using dictionary learning. Instead of posterior values obtained by a Gaussian mixture as in other methods, we use sparse weight vectors and their variants as an alternative noisy speech feature representation. This paper also provides an efficient algorithm that can be applied to large-scale speech processing. We show the effectiveness of the proposed approach by using a middle vocabulary noisy speech recognition task based on WSJ, which was provided by the 2nd CHiME Speech Separation and Recognition Challenge.

KW - 2nd CHiME challenge track 2

KW - dictionary learning

KW - sparse representation

KW - speech feature enhancement

KW - Speech recognition

UR - http://www.scopus.com/inward/record.url?scp=84890470592&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84890470592&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2013.6639034

DO - 10.1109/ICASSP.2013.6639034

M3 - Conference contribution

AN - SCOPUS:84890470592

SN - 9781479903566

SP - 7073

EP - 7077

BT - 2013 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2013 - Proceedings

ER -