The incorporation of masking threshold to subspace speech enhancement

Author

Jong Uk Kim ; Kim, Sang G. ; Yoo, Chang D.

Author_Institution

Dept. of Electr. Eng. & Comput. Sci., Korea Adv. Inst. of Sci. & Technol., Daejon, South Korea

Volume

1

fYear

2003

fDate

6-10 April 2003

Abstract

A subspace enhancement algorithm using the masking property is proposed. The proposed algorithm minimizes the signal distortion while constraining the energy of the residual noise below the human psychoacoustic masking threshold. This requires simple transformation of the masking threshold from the Fourier domain to the Karhunen-Loe`ve (KL) domain. The proposed method incorporates subband whitening filters in the KL domain in order to deal with colored noise. Performance test results show that the proposed algorithm is superior to spectral subtraction and the subspace method suggested by Y. Ephraim and H.L. Van Trees (see IEEE Trans. Speech and Audio Processing, vol.3, no.4, p.251-66, 1995).

Keywords

Fourier transforms; Karhunen-Loeve transforms; distortion; filtering theory; hearing; minimisation; random noise; speech enhancement; Fourier domain; Karhunen-Loeve domain; colored noise; human psychoacoustic masking threshold; residual noise; signal distortion minimization; spectral subtraction; subband whitening filters; subspace speech enhancement; Additive white noise; Covariance matrix; Distortion; Guidelines; Humans; Masking threshold; Noise reduction; Psychology; Speech enhancement; Wiener filter;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 2003. Proceedings. (ICASSP '03). 2003 IEEE International Conference on

ISSN

1520-6149

Print_ISBN

0-7803-7663-3

Type

conf

DOI

10.1109/ICASSP.2003.1198720

Filename

1198720