A novel robust feature of speech signal based on the Mellin transform for speaker-independent speech recognition

Jingdong Chen; Bo Xu; Taiyi Huang

doi:10.1109/ICASSP.1998.675343

A novel robust feature of speech signal based on the Mellin transform for speaker-independent speech recognition

Jingdong Chen
, Bo Xu
, Taiyi Huang

CAS - Institute of Automation

科研成果: 书/报告/会议事项章节 › 会议稿件 › 同行评审

9 引用（Scopus）

摘要

This paper presents a novel kind of speech feature which is the modified Mellin transform of the log-spectrum of the speech signal (short for MMTLS). Because of the scale invariance property of the modified Mellin transform, the new feature is insensitive to the variation of the vocal tract length among individual speakers, and thus it is more appropriate for speaker-independent speech recognition than the popular used cepstrum. The preliminary experiments show that the performance of the MMTLS-based method is much better in comparison with those of the LPC- and MFC-based methods. Moreover, the error rate of this method is very consistent for different outlier speakers.

源语言	英语
主期刊名	Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998
页	629-632
页数	4
DOI	https://doi.org/10.1109/ICASSP.1998.675343
出版状态	已出版 - 1998
已对外发布	是
活动	1998 23rd IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998 - Seattle, WA, 美国期限: 12 5月 1998 → 15 5月 1998

出版系列

姓名	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
卷	2
ISSN（印刷版）	1520-6149

会议

会议	1998 23rd IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998
国家/地区	美国
市	Seattle, WA
时期	12/05/98 → 15/05/98

访问文件

10.1109/ICASSP.1998.675343

其它文件与链接

链接到 Scopus 的出版物

引用此

Chen, J., Xu, B., & Huang, T. (1998). A novel robust feature of speech signal based on the Mellin transform for speaker-independent speech recognition. 在 Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998 (页码 629-632). 文章 675343 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; 卷 2). https://doi.org/10.1109/ICASSP.1998.675343

Chen, Jingdong ; Xu, Bo ; Huang, Taiyi. / A novel robust feature of speech signal based on the Mellin transform for speaker-independent speech recognition. Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998. 1998. 页码 629-632 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).

@inproceedings{e26a1e7f49e14942a864ee7bb97993d7,

title = "A novel robust feature of speech signal based on the Mellin transform for speaker-independent speech recognition",

abstract = "This paper presents a novel kind of speech feature which is the modified Mellin transform of the log-spectrum of the speech signal (short for MMTLS). Because of the scale invariance property of the modified Mellin transform, the new feature is insensitive to the variation of the vocal tract length among individual speakers, and thus it is more appropriate for speaker-independent speech recognition than the popular used cepstrum. The preliminary experiments show that the performance of the MMTLS-based method is much better in comparison with those of the LPC- and MFC-based methods. Moreover, the error rate of this method is very consistent for different outlier speakers.",

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

year = "1998",

doi = "10.1109/ICASSP.1998.675343",

language = "英语",

isbn = "0780344286",

series = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

pages = "629--632",

booktitle = "Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998",

note = "1998 23rd IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998 ; Conference date: 12-05-1998 Through 15-05-1998",

}

Chen, J, Xu, B & Huang, T 1998, A novel robust feature of speech signal based on the Mellin transform for speaker-independent speech recognition. 在 Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998., 675343, ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, 卷 2, 页码 629-632, 1998 23rd IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998, Seattle, WA, 美国, 12/05/98. https://doi.org/10.1109/ICASSP.1998.675343

A novel robust feature of speech signal based on the Mellin transform for speaker-independent speech recognition. / Chen, Jingdong; Xu, Bo; Huang, Taiyi.
Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998. 1998. 页码 629-632 675343 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; 卷 2).

科研成果: 书/报告/会议事项章节 › 会议稿件 › 同行评审

TY - GEN

T1 - A novel robust feature of speech signal based on the Mellin transform for speaker-independent speech recognition

AU - Chen, Jingdong

AU - Xu, Bo

AU - Huang, Taiyi

PY - 1998

Y1 - 1998

N2 - This paper presents a novel kind of speech feature which is the modified Mellin transform of the log-spectrum of the speech signal (short for MMTLS). Because of the scale invariance property of the modified Mellin transform, the new feature is insensitive to the variation of the vocal tract length among individual speakers, and thus it is more appropriate for speaker-independent speech recognition than the popular used cepstrum. The preliminary experiments show that the performance of the MMTLS-based method is much better in comparison with those of the LPC- and MFC-based methods. Moreover, the error rate of this method is very consistent for different outlier speakers.

AB - This paper presents a novel kind of speech feature which is the modified Mellin transform of the log-spectrum of the speech signal (short for MMTLS). Because of the scale invariance property of the modified Mellin transform, the new feature is insensitive to the variation of the vocal tract length among individual speakers, and thus it is more appropriate for speaker-independent speech recognition than the popular used cepstrum. The preliminary experiments show that the performance of the MMTLS-based method is much better in comparison with those of the LPC- and MFC-based methods. Moreover, the error rate of this method is very consistent for different outlier speakers.

UR - https://www.scopus.com/pages/publications/0011498037

U2 - 10.1109/ICASSP.1998.675343

DO - 10.1109/ICASSP.1998.675343

M3 - 会议稿件

AN - SCOPUS:0011498037

SN - 0780344286

SN - 9780780344280

T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

SP - 629

EP - 632

BT - Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998

T2 - 1998 23rd IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998

Y2 - 12 May 1998 through 15 May 1998

ER -

Chen J, Xu B, Huang T. A novel robust feature of speech signal based on the Mellin transform for speaker-independent speech recognition. 在 Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998. 1998. 页码 629-632. 675343. (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). doi: 10.1109/ICASSP.1998.675343

A novel robust feature of speech signal based on the Mellin transform for speaker-independent speech recognition

摘要

出版系列

会议

访问文件

其它文件与链接

指纹

引用此