Fast Noise Suppression Algorithm with Kalman Filter Theory

Author

Tanabe, Nari ; Furukawa, Toshihiro ; Tsujii, Shigeo

Author_Institution

Tokyo Univ. of Sci., Chino

fYear

2008

fDate

15-16 Dec. 2008

Firstpage

411

Lastpage

415

Abstract

We have proposed a robust noise suppression algorithm with Kalman filter theory. In this paper, we propose a Kalman filter based fast noise suppression algorithm for white and colored disturbance. The algorithm aims to achieve robust noise suppression with reduced computational complexity without sacrificing high quality of speech signal, by modifying the proposed canonical space model. We show the effectiveness of the proposed fast noise suppression algorithm using numerical and subjective evaluation results.

Keywords

Kalman filters; computational complexity; interference suppression; speech processing; white noise; Kalman filter theory; canonical space model; colored disturbance; computational complexity; fast noise suppression algorithm; speech signal quality; white disturbance; Additive noise; Colored noise; Computational complexity; Equations; Noise generators; Noise reduction; Noise robustness; Signal generators; Speech enhancement; State-space methods; canonical state space models; colored driving source; computational complexity; fast robust noise suppression; kalman filter theory; white and colored noise;

fLanguage

English

Publisher

ieee

Conference_Titel

Universal Communication, 2008. ISUC '08. Second International Symposium on

Conference_Location

Osaka

Print_ISBN

978-0-7695-3433-6

Type

conf

DOI

10.1109/ISUC.2008.34

Filename

4724494