DocumentCode :
699443
Title :
Robust spectrum quantization for LP parameter enhancement
Author :
Grancharov, Volodya ; Srinivasan, Sriram ; Samuelsson, Jonas ; Kleijn, W. Bastiaan
Author_Institution :
Dept. of Signals, Sensors & Syst., KTH (R. Inst. of Technol.), Stockholm, Sweden
fYear :
2004
fDate :
6-10 Sept. 2004
Firstpage :
1951
Lastpage :
1954
Abstract :
In this paper, we investigate the denoising properties of robust vector quantization of the speech spectrum parameters in combination with a Kalman filter. The underlying assumption is that the high-energy speech regions can be used to reconstruct the low-energy regions destroyed by noise. This can be achieved through vector quantization with a properly weighted distortion measure. The performance of the proposed system, Kalman filtering with prior vector quantization, is compared with existing schemes for parameter estimation used in Kalman filtering. The results indicate significant improvement over the reference systems in both objective and subjective tests.
Keywords :
Kalman filters; linear predictive coding; signal denoising; speech enhancement; vector quantisation; Kalman filter; LP parameter enhancement; high-energy speech region; linear prediction coefficients; speech spectrum parameter robust vector quantization denoising properties; weighted distortion measure; Abstracts; Noise; Quantization (signal); Robustness; Speech; Speech coding; Speech enhancement;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Signal Processing Conference, 2004 12th European
Conference_Location :
Vienna
Print_ISBN :
978-320-0001-65-7
Type :
conf
Filename :
7079973
Link To Document :
بازگشت