A new group delay-based feature for robust speech recognition

E. Loweimi; Seyed Mohammad Ahadi

doi:10.1109/ICME.2011.6011884

A new group delay-based feature for robust speech recognition

E. Loweimi, Seyed Mohammad Ahadi

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

Abstract / Description of output

In this paper we present a novel feature extraction algorithm based on group delay function for robust speech recognition. The modified group delay function (MODGDF) is the main feature extraction method based on group delay function, generally used for robust speech recognition. The recognition tests indicate this feature does not provide notably better results in the presence of additive noise in comparison with MFCC. In the presence of convolutional noise, the performance of MODGDF is considerably lower than MFCC. The method proposed in this paper is simple and makes more efficient utilization of the high resolution property of GDF. It is formed from three main parts which are signal modeling, GDF computation based on extracted model, and compression. The recognition results obtained over AURORA 2.0 task indicate its superior performance in comparison with MODGDF and MFCC.

Original language	English
Title of host publication	2011 IEEE International Conference on Multimedia and Expo
Publisher	Institute of Electrical and Electronics Engineers
Pages	1-5
Number of pages	5
ISBN (Electronic)	978-1-61284-349-0
ISBN (Print)	978-1-61284-348-3
DOIs	https://doi.org/10.1109/ICME.2011.6011884
Publication status	Published - 1 Jul 2011
Event	2011 IEEE International Conference on Multimedia and Expo - Barcelona, Spain Duration: 11 Jul 2011 → 15 Jul 2011 http://www.ieee-icme.org/icme2011/

Conference

Conference	2011 IEEE International Conference on Multimedia and Expo
Abbreviated title	ICME 2011
Country/Territory	Spain
City	Barcelona
Period	11/07/11 → 15/07/11
Internet address	http://www.ieee-icme.org/icme2011/

Keywords / Materials (for Non-textual outputs)

Delay
Speech
Mel frequency cepstral coefficient
Speech recognition
Feature extraction
Computational modeling
Discrete cosine transforms
Robust speech recognition
group delay function
signal modeling
compression

Access to Document

10.1109/ICME.2011.6011884

https://ieeexplore.ieee.org/document/6011884

Cite this

@inproceedings{e9f25e333b2c4ea6b3e853f6e37a6002,

title = "A new group delay-based feature for robust speech recognition",

abstract = "In this paper we present a novel feature extraction algorithm based on group delay function for robust speech recognition. The modified group delay function (MODGDF) is the main feature extraction method based on group delay function, generally used for robust speech recognition. The recognition tests indicate this feature does not provide notably better results in the presence of additive noise in comparison with MFCC. In the presence of convolutional noise, the performance of MODGDF is considerably lower than MFCC. The method proposed in this paper is simple and makes more efficient utilization of the high resolution property of GDF. It is formed from three main parts which are signal modeling, GDF computation based on extracted model, and compression. The recognition results obtained over AURORA 2.0 task indicate its superior performance in comparison with MODGDF and MFCC.",

keywords = "Delay, Speech, Mel frequency cepstral coefficient, Speech recognition, Feature extraction, Computational modeling, Discrete cosine transforms, Robust speech recognition, group delay function, signal modeling, compression",

author = "E. Loweimi and Ahadi, {Seyed Mohammad}",

year = "2011",

month = jul,

day = "1",

doi = "10.1109/ICME.2011.6011884",

language = "English",

isbn = "978-1-61284-348-3",

pages = "1--5",

booktitle = "2011 IEEE International Conference on Multimedia and Expo",

publisher = "Institute of Electrical and Electronics Engineers",

address = "United States",

note = "2011 IEEE International Conference on Multimedia and Expo, ICME 2011 ; Conference date: 11-07-2011 Through 15-07-2011",

url = "http://www.ieee-icme.org/icme2011/",

}

TY - GEN

T1 - A new group delay-based feature for robust speech recognition

AU - Loweimi, E.

AU - Ahadi, Seyed Mohammad

PY - 2011/7/1

Y1 - 2011/7/1

N2 - In this paper we present a novel feature extraction algorithm based on group delay function for robust speech recognition. The modified group delay function (MODGDF) is the main feature extraction method based on group delay function, generally used for robust speech recognition. The recognition tests indicate this feature does not provide notably better results in the presence of additive noise in comparison with MFCC. In the presence of convolutional noise, the performance of MODGDF is considerably lower than MFCC. The method proposed in this paper is simple and makes more efficient utilization of the high resolution property of GDF. It is formed from three main parts which are signal modeling, GDF computation based on extracted model, and compression. The recognition results obtained over AURORA 2.0 task indicate its superior performance in comparison with MODGDF and MFCC.

AB - In this paper we present a novel feature extraction algorithm based on group delay function for robust speech recognition. The modified group delay function (MODGDF) is the main feature extraction method based on group delay function, generally used for robust speech recognition. The recognition tests indicate this feature does not provide notably better results in the presence of additive noise in comparison with MFCC. In the presence of convolutional noise, the performance of MODGDF is considerably lower than MFCC. The method proposed in this paper is simple and makes more efficient utilization of the high resolution property of GDF. It is formed from three main parts which are signal modeling, GDF computation based on extracted model, and compression. The recognition results obtained over AURORA 2.0 task indicate its superior performance in comparison with MODGDF and MFCC.

KW - Delay

KW - Speech

KW - Mel frequency cepstral coefficient

KW - Speech recognition

KW - Feature extraction

KW - Computational modeling

KW - Discrete cosine transforms

KW - Robust speech recognition

KW - group delay function

KW - signal modeling

KW - compression

U2 - 10.1109/ICME.2011.6011884

DO - 10.1109/ICME.2011.6011884

M3 - Conference contribution

SN - 978-1-61284-348-3

SP - 1

EP - 5

BT - 2011 IEEE International Conference on Multimedia and Expo

PB - Institute of Electrical and Electronics Engineers

T2 - 2011 IEEE International Conference on Multimedia and Expo

Y2 - 11 July 2011 through 15 July 2011

ER -

A new group delay-based feature for robust speech recognition

Abstract / Description of output

Conference

Keywords / Materials (for Non-textual outputs)

Access to Document

Fingerprint

Cite this