Template-based Spectral Estimation using microphone array for speech recognition

Satoshi Tamura; Eriko Hishikawa; Wataru Taguchi; Satoru Hayamizu

Template-based Spectral Estimation using microphone array for speech recognition

Satoshi Tamura^*, Eriko Hishikawa, Wataru Taguchi, Satoru Hayamizu

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

This paper proposes a Template-based Spectral Estimation (TSE) method for noise reduction of microphone array processing aiming at speech recognition enhancement. In the proposed method, a noise template in a complex plane is calculated for each frequency bin using non-speech audio signals observed at microphones. Then for every noise-overlapped speech signals, a speech signal can be reformed by applying the template and the gradient descent method. Experiments were conducted to evaluate not only performance of noise reduction but also improvement of speech recognition. Then NRR 16.7dB improvement was achieved by combining TSE and Spectral Subtraction (SS) methods. For speech recognition, 44% relative recognition error reduction was obtained comparing with the conventional SS method.

Original language	English
Title of host publication	Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010
Publisher	International Speech Communication Association
Pages	2050-2053
Number of pages	4
Publication status	Published - 2010
Externally published	Yes

Publication series

Name	Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010

Keywords

Microphone array
Noise reduction
Spectral sub-truction
Speech recognition enhancement

ASJC Scopus subject areas

Language and Linguistics
Speech and Hearing
Human-Computer Interaction
Signal Processing
Software
Modelling and Simulation

Cite this

Tamura, S., Hishikawa, E., Taguchi, W., & Hayamizu, S. (2010). Template-based Spectral Estimation using microphone array for speech recognition. In Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010 (pp. 2050-2053). (Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010). International Speech Communication Association.

Template-based Spectral Estimation using microphone array for speech recognition. / Tamura, Satoshi; Hishikawa, Eriko; Taguchi, Wataru et al.
Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010. International Speech Communication Association, 2010. p. 2050-2053 (Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Tamura, S, Hishikawa, E, Taguchi, W & Hayamizu, S 2010, Template-based Spectral Estimation using microphone array for speech recognition. in Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010. Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010, International Speech Communication Association, pp. 2050-2053.

Tamura S, Hishikawa E, Taguchi W, Hayamizu S. Template-based Spectral Estimation using microphone array for speech recognition. In Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010. International Speech Communication Association. 2010. p. 2050-2053. (Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010).

Tamura, Satoshi ; Hishikawa, Eriko ; Taguchi, Wataru et al. / Template-based Spectral Estimation using microphone array for speech recognition. Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010. International Speech Communication Association, 2010. pp. 2050-2053 (Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010).

@inproceedings{2261eb3280784742be9f86a070df6d2e,

title = "Template-based Spectral Estimation using microphone array for speech recognition",

abstract = "This paper proposes a Template-based Spectral Estimation (TSE) method for noise reduction of microphone array processing aiming at speech recognition enhancement. In the proposed method, a noise template in a complex plane is calculated for each frequency bin using non-speech audio signals observed at microphones. Then for every noise-overlapped speech signals, a speech signal can be reformed by applying the template and the gradient descent method. Experiments were conducted to evaluate not only performance of noise reduction but also improvement of speech recognition. Then NRR 16.7dB improvement was achieved by combining TSE and Spectral Subtraction (SS) methods. For speech recognition, 44% relative recognition error reduction was obtained comparing with the conventional SS method.",

keywords = "Microphone array, Noise reduction, Spectral sub-truction, Speech recognition enhancement",

author = "Satoshi Tamura and Eriko Hishikawa and Wataru Taguchi and Satoru Hayamizu",

year = "2010",

language = "English",

series = "Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010",

publisher = "International Speech Communication Association",

pages = "2050--2053",

booktitle = "Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010",

}

TY - GEN

T1 - Template-based Spectral Estimation using microphone array for speech recognition

AU - Tamura, Satoshi

AU - Hishikawa, Eriko

AU - Taguchi, Wataru

AU - Hayamizu, Satoru

PY - 2010

Y1 - 2010

N2 - This paper proposes a Template-based Spectral Estimation (TSE) method for noise reduction of microphone array processing aiming at speech recognition enhancement. In the proposed method, a noise template in a complex plane is calculated for each frequency bin using non-speech audio signals observed at microphones. Then for every noise-overlapped speech signals, a speech signal can be reformed by applying the template and the gradient descent method. Experiments were conducted to evaluate not only performance of noise reduction but also improvement of speech recognition. Then NRR 16.7dB improvement was achieved by combining TSE and Spectral Subtraction (SS) methods. For speech recognition, 44% relative recognition error reduction was obtained comparing with the conventional SS method.

AB - This paper proposes a Template-based Spectral Estimation (TSE) method for noise reduction of microphone array processing aiming at speech recognition enhancement. In the proposed method, a noise template in a complex plane is calculated for each frequency bin using non-speech audio signals observed at microphones. Then for every noise-overlapped speech signals, a speech signal can be reformed by applying the template and the gradient descent method. Experiments were conducted to evaluate not only performance of noise reduction but also improvement of speech recognition. Then NRR 16.7dB improvement was achieved by combining TSE and Spectral Subtraction (SS) methods. For speech recognition, 44% relative recognition error reduction was obtained comparing with the conventional SS method.

KW - Microphone array

KW - Noise reduction

KW - Spectral sub-truction

KW - Speech recognition enhancement

UR - http://www.scopus.com/inward/record.url?scp=79959832425&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79959832425&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:79959832425

T3 - Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010

SP - 2050

EP - 2053

BT - Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010

PB - International Speech Communication Association

ER -

Template-based Spectral Estimation using microphone array for speech recognition

Abstract

Publication series

Keywords

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this