Markov model based noise modeling and its application to noisy speech recognition using dynamical features of speech

Tetsunori Kobayashi; Ryuji Mine; Katsuhiko Shirai

doi:10.1109/ICASSP.1994.389719

Markov model based noise modeling and its application to noisy speech recognition using dynamical features of speech

Tetsunori Kobayashi^*, Ryuji Mine, Katsuhiko Shirai

^*Corresponding author for this work

School of Fundamental Science and Engineering

Research output: Contribution to journal › Conference article › peer-review

3 Citations (Scopus)

Abstract

In this paper, some algorithms to recognize speech in time varying noise are proposed. In the proposed methods, Spectral subtraction and Markov model based noise models are successfully utilized in the framework of spectral decomposition of noisy speech. Firstly, we considered the problem of the mis-subtraction noise which is caused in the subtraction based decomposition procedure. Then, the precise use of dynamical feature of speech such as delta cepstrum is discussed. Using the methods proposed here, recognition performance are improved more than 60 % compared to no compensation method.

Original language	English
Article number	389719
Pages (from-to)	II57-II60
Journal	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume	2
DOIs	https://doi.org/10.1109/ICASSP.1994.389719
Publication status	Published - 1994
Event	Proceedings of the 1994 IEEE International Conference on Acoustics, Speech and Signal Processing. Part 2 (of 6) - Adelaide, Aust Duration: 1994 Apr 19 → 1994 Apr 22

ASJC Scopus subject areas

Software
Signal Processing
Electrical and Electronic Engineering

Access to Document

10.1109/ICASSP.1994.389719

Cite this

Markov model based noise modeling and its application to noisy speech recognition using dynamical features of speech. / Kobayashi, Tetsunori; Mine, Ryuji; Shirai, Katsuhiko.
In: ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, Vol. 2, 389719, 1994, p. II57-II60.

Research output: Contribution to journal › Conference article › peer-review

@article{af1d7427e7bd41bda2a10effe8b71053,

title = "Markov model based noise modeling and its application to noisy speech recognition using dynamical features of speech",

abstract = "In this paper, some algorithms to recognize speech in time varying noise are proposed. In the proposed methods, Spectral subtraction and Markov model based noise models are successfully utilized in the framework of spectral decomposition of noisy speech. Firstly, we considered the problem of the mis-subtraction noise which is caused in the subtraction based decomposition procedure. Then, the precise use of dynamical feature of speech such as delta cepstrum is discussed. Using the methods proposed here, recognition performance are improved more than 60 % compared to no compensation method.",

author = "Tetsunori Kobayashi and Ryuji Mine and Katsuhiko Shirai",

year = "1994",

doi = "10.1109/ICASSP.1994.389719",

language = "English",

volume = "2",

pages = "II57--II60",

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

issn = "0736-7791",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

note = "Proceedings of the 1994 IEEE International Conference on Acoustics, Speech and Signal Processing. Part 2 (of 6) ; Conference date: 19-04-1994 Through 22-04-1994",

}

TY - JOUR

T1 - Markov model based noise modeling and its application to noisy speech recognition using dynamical features of speech

AU - Kobayashi, Tetsunori

AU - Mine, Ryuji

AU - Shirai, Katsuhiko

PY - 1994

Y1 - 1994

N2 - In this paper, some algorithms to recognize speech in time varying noise are proposed. In the proposed methods, Spectral subtraction and Markov model based noise models are successfully utilized in the framework of spectral decomposition of noisy speech. Firstly, we considered the problem of the mis-subtraction noise which is caused in the subtraction based decomposition procedure. Then, the precise use of dynamical feature of speech such as delta cepstrum is discussed. Using the methods proposed here, recognition performance are improved more than 60 % compared to no compensation method.

AB - In this paper, some algorithms to recognize speech in time varying noise are proposed. In the proposed methods, Spectral subtraction and Markov model based noise models are successfully utilized in the framework of spectral decomposition of noisy speech. Firstly, we considered the problem of the mis-subtraction noise which is caused in the subtraction based decomposition procedure. Then, the precise use of dynamical feature of speech such as delta cepstrum is discussed. Using the methods proposed here, recognition performance are improved more than 60 % compared to no compensation method.

UR - http://www.scopus.com/inward/record.url?scp=5844294232&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=5844294232&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.1994.389719

DO - 10.1109/ICASSP.1994.389719

M3 - Conference article

AN - SCOPUS:5844294232

SN - 0736-7791

VL - 2

SP - II57-II60

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

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

M1 - 389719

T2 - Proceedings of the 1994 IEEE International Conference on Acoustics, Speech and Signal Processing. Part 2 (of 6)

Y2 - 19 April 1994 through 22 April 1994

ER -

Markov model based noise modeling and its application to noisy speech recognition using dynamical features of speech

Abstract

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this