Presentation 2014-06-19
Instantaneous amplitude and phase restoration using Kalman filter for speech enhancement
Yang LIU, Naushin NOWER, Masashi UNOKI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper proposes a method of restoring the instantaneous amplitudes and phases using Kalman filter with linear prediction from a noisy reverberant signal for speech enhancement. Recent studies have shown that the phase spectrum in short time Fourier transform plays an important role for speech enhancement. We thus have previously proposed the method of restoring the instantaneous amplitudes and phases simultaneously in noisy environment based on this fact. In this paper, we extend our previous method to be general speech enhancement in real environments by removing the effects of noise and reverberation simultaneously. Objective experiments were conducted in various noisy reverberant environments to evaluate the effectiveness of the proposed scheme by using signal to error ratio (SER), correlation, perceptual evaluation of speech quality (PESQ), and SNR loss. Results showed that the proposed method can effectively improve these objective measurements in noisy reverberant environments, compared with typical methods such as Wiener filtering method and minimum mean square error short-time spectral amplitude (MMSE-STSA) method.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Instantaneous amplitude and phase / Kalman filter / Linear prediction / Gammatone filterbank
Paper # SP2014-51,WIT2014-6
Date of Issue

Conference Information
Committee WIT
Conference Date 2014/6/12(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 Well-being Information Technology(WIT)
Language ENG
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Instantaneous amplitude and phase restoration using Kalman filter for speech enhancement
Sub Title (in English)
Keyword(1) Instantaneous amplitude and phase
Keyword(2) Kalman filter
Keyword(3) Linear prediction
Keyword(4) Gammatone filterbank
1st Author's Name Yang LIU
1st Author's Affiliation School of Information Science, Japan Advanced Institute of Science and Technology,()
2nd Author's Name Naushin NOWER
2nd Author's Affiliation School of Information Science, Japan Advanced Institute of Science and Technology,
3rd Author's Name Masashi UNOKI
3rd Author's Affiliation School of Information Science, Japan Advanced Institute of Science and Technology,
Date 2014-06-19
Paper # SP2014-51,WIT2014-6
Volume (vol) vol.114
Number (no) 92
Page pp.pp.-
#Pages 6
Date of Issue