A Noise Reduction Method Based on Linear Prediction with Variable Step-Size

Arata KAWAMURA; Youji IIGUNI; Yoshio ITOH

doi:10.1093/ietfec/e88-a.4.855

A Noise Reduction Method Based on Linear Prediction with Variable Step-Size

Arata KAWAMURA, Youji IIGUNI, Yoshio ITOH

Full Text Views

0

Share
Cite this

Summary :

A noise reduction technique that uses the linear prediction to remove noise components in speech signals has been proposed previously. The noise reduction works well for additive white noise signals, because the coefficients of the linear predictor converge such that the prediction error becomes white. In this method, the linear predictor is updated by a gradient-based algorithm with a fixed step-size. However, the optimal value of the step-size changes with the values of the prediction coefficients. In this paper, we propose a noise reduction system using the linear predictor with a variable step-size. The optimal value of the step-size depends also on the variance of the white noise, however the variance is unknown. We therefore introduce a speech/non-speech detector, and estimate the variance in non-speech segments where the observed signal includes only noise components. The simulation results show that the noise reduction capability of the proposed system is better than that of the conventional one with a fixed step-size.

Publication: IEICE TRANSACTIONS on Fundamentals Vol.E88-A No.4 pp.855-861

Publication Date: 2005/04/01

Publicized

Online ISSN

DOI: 10.1093/ietfec/e88-a.4.855

Type of Manuscript: Special Section PAPER (Special Section on Selected Papers from the 17th Workshop on Circuits and Systems in Karuizawa)

Category

Cite this

Copy

Arata KAWAMURA, Youji IIGUNI, Yoshio ITOH, "A Noise Reduction Method Based on Linear Prediction with Variable Step-Size" in IEICE TRANSACTIONS on Fundamentals, vol. E88-A, no. 4, pp. 855-861, April 2005, doi: 10.1093/ietfec/e88-a.4.855.
Abstract: A noise reduction technique that uses the linear prediction to remove noise components in speech signals has been proposed previously. The noise reduction works well for additive white noise signals, because the coefficients of the linear predictor converge such that the prediction error becomes white. In this method, the linear predictor is updated by a gradient-based algorithm with a fixed step-size. However, the optimal value of the step-size changes with the values of the prediction coefficients. In this paper, we propose a noise reduction system using the linear predictor with a variable step-size. The optimal value of the step-size depends also on the variance of the white noise, however the variance is unknown. We therefore introduce a speech/non-speech detector, and estimate the variance in non-speech segments where the observed signal includes only noise components. The simulation results show that the noise reduction capability of the proposed system is better than that of the conventional one with a fixed step-size.
URL: https://globals.ieice.org/en_transactions/fundamentals/10.1093/ietfec/e88-a.4.855/_p

Copy

@ARTICLE{e88-a_4_855,
author={Arata KAWAMURA, Youji IIGUNI, Yoshio ITOH, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={A Noise Reduction Method Based on Linear Prediction with Variable Step-Size},
year={2005},
volume={E88-A},
number={4},
pages={855-861},
abstract={A noise reduction technique that uses the linear prediction to remove noise components in speech signals has been proposed previously. The noise reduction works well for additive white noise signals, because the coefficients of the linear predictor converge such that the prediction error becomes white. In this method, the linear predictor is updated by a gradient-based algorithm with a fixed step-size. However, the optimal value of the step-size changes with the values of the prediction coefficients. In this paper, we propose a noise reduction system using the linear predictor with a variable step-size. The optimal value of the step-size depends also on the variance of the white noise, however the variance is unknown. We therefore introduce a speech/non-speech detector, and estimate the variance in non-speech segments where the observed signal includes only noise components. The simulation results show that the noise reduction capability of the proposed system is better than that of the conventional one with a fixed step-size.},
keywords={},
doi={10.1093/ietfec/e88-a.4.855},
ISSN={},
month={April},}

Copy

TY - JOUR
TI - A Noise Reduction Method Based on Linear Prediction with Variable Step-Size
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 855
EP - 861
AU - Arata KAWAMURA
AU - Youji IIGUNI
AU - Yoshio ITOH
PY - 2005
DO - 10.1093/ietfec/e88-a.4.855
JO - IEICE TRANSACTIONS on Fundamentals
SN -
VL - E88-A
IS - 4
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - April 2005
AB - A noise reduction technique that uses the linear prediction to remove noise components in speech signals has been proposed previously. The noise reduction works well for additive white noise signals, because the coefficients of the linear predictor converge such that the prediction error becomes white. In this method, the linear predictor is updated by a gradient-based algorithm with a fixed step-size. However, the optimal value of the step-size changes with the values of the prediction coefficients. In this paper, we propose a noise reduction system using the linear predictor with a variable step-size. The optimal value of the step-size depends also on the variance of the white noise, however the variance is unknown. We therefore introduce a speech/non-speech detector, and estimate the variance in non-speech segments where the observed signal includes only noise components. The simulation results show that the noise reduction capability of the proposed system is better than that of the conventional one with a fixed step-size.
ER -