Title :
Nonstationary noise estimation using low-frequency regions for spectral subtraction
Author :
Yamashita, Kohei ; Shimamura, Tetsuya
Author_Institution :
Graduate Sch. of Sci. & Eng., Saitama Univ., Japan
fDate :
6/1/2005 12:00:00 AM
Abstract :
In this letter, a noise estimation method for spectral subtraction is proposed by using low-frequency regions of noisy speech. The method allows tracking the time variation of noise without complicated computation. The performance of a spectral subtraction method based on the new noise estimation is investigated, and its effectiveness is shown in practical experiments.
Keywords :
estimation theory; spectral analysis; speech enhancement; low-frequency region; noise estimation method; noisy speech; spectral subtraction method; speech enhancement; Degradation; Equations; Fourier transforms; Low-frequency noise; Oral communication; Sampling methods; Speech analysis; Speech enhancement; Speech processing; Speech recognition; Low-frequency regions; noise estimation; nonstationary noise; spectral subtraction; speech enhancement;
Journal_Title :
Signal Processing Letters, IEEE
DOI :
10.1109/LSP.2005.847864
