Multiband Excitation for Speech Enhancement

Author

Charoenruengkit, Werayuth ; Erdol, Nurgun

fYear

2009

fDate

4-7 Jan. 2009

Firstpage

10

Lastpage

15

Abstract

The speech enhancement algorithm proposed aims to improve the quality of denoised speech by introducing voicing information to a Wiener-type spectral amplitude gain function. A constrained multiband excitation (MBE) model is used to emphasize harmonic components of the glottal input; and a low variance and bias autoregressive multitaper (ARMT) estimate models the vocal tract. Objective and subjective evaluations show improvement over unconstrained models and those using high variance spectrum estimators.

Keywords

speech enhancement; Wiener-type spectral amplitude gain function; bias autoregressive multitaper estimate models; constrained multiband excitation model; harmonic components; speech enhancement algorithm; variance spectrum estimators; vocal tract; Background noise; Cepstral analysis; Frequency estimation; Noise reduction; Power harmonic filters; Signal processing; Signal to noise ratio; Speech analysis; Speech enhancement; Speech processing; Spectral estimation; Speech communication; Speech enhancement; Speech processing;

fLanguage

English

Publisher

ieee

Conference_Titel

Digital Signal Processing Workshop and 5th IEEE Signal Processing Education Workshop, 2009. DSP/SPE 2009. IEEE 13th

Conference_Location

Marco Island, FL

Print_ISBN

978-1-4244-3677-4

Electronic_ISBN

978-1-4244-3677-4

Type

conf

DOI

10.1109/DSP.2009.4785887

Filename

4785887