An improved log-spectral amplitude estimator algorithm for speech enhancement

Author

Yanfen Wang ; Shouhao Wu ; Wenxiu Guo ; Jianhua Ji

Author_Institution

Res. Inst., Tsinghua Univ. in Shenzhen, Shenzhen, China

fYear

2014

fDate

25-27 June 2014

Firstpage

1

Lastpage

6

Abstract

An improved speech enhancement scheme is proposed based on log-spectral amplitude minimum mean-square error (LSA-MMSE) method in this paper. Burg spectrum and speech absent probability (SAP) are used to improve estimation of priori signal-to-noise ratios (SNR). The estimation of priori SNR is based on "decision-directed" criterion. Burg spectrum is introduced for noisy speech power spectrum estimation. The SAP is utilized to adjust smoothing parameter. Simulation results show that the improved estimator can well balance de-noising degree, residual musical noise and voice distortion, making enhanced voice sounds more natural.

Keywords

mean square error methods; probability; speech enhancement; Burg spectrum; LSA-MMSE method; SAP; SNR; decision directed criterion; denoising degree; improved log spectral amplitude estimator algorithm; log spectral amplitude minimum mean square error; noisy speech power spectrum estimation; residual musical noise; signal-to-noise ratios; smoothing parameter; speech absent probability; speech enhancement scheme; voice distortion; voice sounds; Estimation; Noise measurement; Signal to noise ratio; Smoothing methods; Speech; Speech enhancement; Audio technology; Log-spectral Amplitude Estimator; Multimedia processing; Speech Enhancement;

fLanguage

English

Publisher

ieee

Conference_Titel

Broadband Multimedia Systems and Broadcasting (BMSB), 2014 IEEE International Symposium on

Conference_Location

Beijing

Type

conf

DOI

10.1109/BMSB.2014.6873490

Filename

6873490