Speech enhancement based on audible noise suppression

Author

Tsoukalas, Dionysis E. ; Mourjopoulos, John N. ; Kokkinakis, George

Author_Institution

Wire Commun. Lab., Patras Univ., Greece

Volume

5

Issue

6

fYear

1997

fDate

11/1/1997 12:00:00 AM

Firstpage

497

Lastpage

514

Abstract

A novel speech enhancement technique is presented based on the definition of the psychoacoustically derived quantity of audible noise spectrum and its subsequent suppression using optimal nonlinear filtering of the short-time spectral amplitude (STSA) envelope. The filter operates with sparse spectral estimates obtained from the STSA, and, when these parameters are accurately known, significant intelligibility gains, up to 40%, result in the processed speech signal. These parameters can be also estimated from noisy data, resulting into smaller but significant intelligibility gains

Keywords

acoustic signal processing; filtering theory; noise; parameter estimation; speech enhancement; speech intelligibility; speech processing; audible noise spectrum; audible noise suppression; intelligibility gains; noisy data; optimal nonlinear filtering; parameter estimation; psychoacoustic model; short-time spectral amplitude envelope; sparse spectral estimates; speech enhancement; speech signal; Acoustic noise; Additive noise; Auditory system; Degradation; Filtering; Filters; Humans; Psychoacoustic models; Psychology; Speech enhancement;

fLanguage

English

Journal_Title

Speech and Audio Processing, IEEE Transactions on

Publisher

ieee

ISSN

1063-6676

Type

jour

DOI

10.1109/89.641296

Filename

641296