New features based on the Cohen's class of bilinear time-frequency representations for speech recognition

Jingdong Chen; Bo Xu; Taiyi Huang

New features based on the Cohen's class of bilinear time-frequency representations for speech recognition

Jingdong Chen, Bo Xu, Taiyi Huang

Chinese Academy of Sciences

科研成果: 会议稿件 › 论文 › 同行评审

摘要

Although short-time Fourier Analysis based features such as LPCC and MFCC have been widely used in state-of-the-art speech recognizers, the short-time analysis technique suffers from the well-known trade-off between time and frequency resolution and works under the assumption that speech signal is short-time stationary. This paper investigates an approach to use Cohen's class bilinear time-frequency distributions representing speech signal for speech recognition. Preliminary experiments show that the new feature can better represent speech signal and can improve the accuracy of a speech recognizer.

源语言	英语
页	674-677
页数	4
出版状态	已出版 - 1998
已对外发布	是
活动	Proceedings of the 1998 4th International Conference on Signal Processing Proceedings, ICSP '98 - Beijing, China 期限: 12 10月 1998 → 16 10月 1998

会议

会议	Proceedings of the 1998 4th International Conference on Signal Processing Proceedings, ICSP '98
市	Beijing, China
时期	12/10/98 → 16/10/98

其它文件与链接

链接到 Scopus 的出版物

引用此

@conference{358a5dc48a8047dfac1794cea36ad752,

title = "New features based on the Cohen's class of bilinear time-frequency representations for speech recognition",

abstract = "Although short-time Fourier Analysis based features such as LPCC and MFCC have been widely used in state-of-the-art speech recognizers, the short-time analysis technique suffers from the well-known trade-off between time and frequency resolution and works under the assumption that speech signal is short-time stationary. This paper investigates an approach to use Cohen's class bilinear time-frequency distributions representing speech signal for speech recognition. Preliminary experiments show that the new feature can better represent speech signal and can improve the accuracy of a speech recognizer.",

author = "Jingdong Chen and Bo Xu and Taiyi Huang",

year = "1998",

language = "英语",

pages = "674--677",

note = "Proceedings of the 1998 4th International Conference on Signal Processing Proceedings, ICSP '98 ; Conference date: 12-10-1998 Through 16-10-1998",

}

TY - CONF

T1 - New features based on the Cohen's class of bilinear time-frequency representations for speech recognition

AU - Chen, Jingdong

AU - Xu, Bo

AU - Huang, Taiyi

PY - 1998

Y1 - 1998

N2 - Although short-time Fourier Analysis based features such as LPCC and MFCC have been widely used in state-of-the-art speech recognizers, the short-time analysis technique suffers from the well-known trade-off between time and frequency resolution and works under the assumption that speech signal is short-time stationary. This paper investigates an approach to use Cohen's class bilinear time-frequency distributions representing speech signal for speech recognition. Preliminary experiments show that the new feature can better represent speech signal and can improve the accuracy of a speech recognizer.

AB - Although short-time Fourier Analysis based features such as LPCC and MFCC have been widely used in state-of-the-art speech recognizers, the short-time analysis technique suffers from the well-known trade-off between time and frequency resolution and works under the assumption that speech signal is short-time stationary. This paper investigates an approach to use Cohen's class bilinear time-frequency distributions representing speech signal for speech recognition. Preliminary experiments show that the new feature can better represent speech signal and can improve the accuracy of a speech recognizer.

UR - http://www.scopus.com/inward/record.url?scp=0032272639&partnerID=8YFLogxK

M3 - 论文

AN - SCOPUS:0032272639

SP - 674

EP - 677

T2 - Proceedings of the 1998 4th International Conference on Signal Processing Proceedings, ICSP '98

Y2 - 12 October 1998 through 16 October 1998

ER -

New features based on the Cohen's class of bilinear time-frequency representations for speech recognition

摘要

会议

其它文件与链接

指纹

引用此