Presentation 2005/7/15
An Improved mel-Wiener Filter for Mel-LPC based Speech Recognition
BABUL ISLAM, Hiroshi MATSUMOTO, Kazumasa YAMAMOTO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper presents a frequency warped Wiener filter to enhance Mel-LPC spectra in presence of additive noise. The proposed filter is estimated based on minimization of error signal on the linear frequency scale and then is efficiently implemented in the autocorrelation domain without denoising input speech. The performance of the proposed filter is evaluated on Aurora 2 database. The optimum filter order is 5, which is shown to be lower than that of Mel-LPC analysis, and thus filtering is computationally inexpensive. As a result, word accuracy is improved by about 49% at most with the proposed Wiener filter.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Wiener filter / Mel-Wiener filter / Mel-LPC analysis / Spectral subtraction / Noisy speech recognition
Paper # SP2005-42
Date of Issue

Conference Information
Committee SP
Conference Date 2005/7/15(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Speech (SP)
Language ENG
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) An Improved mel-Wiener Filter for Mel-LPC based Speech Recognition
Sub Title (in English)
Keyword(1) Wiener filter
Keyword(2) Mel-Wiener filter
Keyword(3) Mel-LPC analysis
Keyword(4) Spectral subtraction
Keyword(5) Noisy speech recognition
1st Author's Name BABUL ISLAM
1st Author's Affiliation Graduate School of Science and Technology, Shinshu University()
2nd Author's Name Hiroshi MATSUMOTO
2nd Author's Affiliation Faculty of Engineering, Shinshu University
3rd Author's Name Kazumasa YAMAMOTO
3rd Author's Affiliation Faculty of Engineering, Shinshu University
Date 2005/7/15
Paper # SP2005-42
Volume (vol) vol.105
Number (no) 199
Page pp.pp.-
#Pages 6
Date of Issue