DocumentCode :
3422595
Title :
Environment-invariant compensation for reverberation using linear post-filtering for minimum distortion
Author :
Kumar, Kshitiz ; Stern, Richard M.
Author_Institution :
Dept. of Electr. & Comput. Eng., Carnegie Mellon Univ., Pittsburgh, PA
fYear :
2008
fDate :
March 31 2008-April 4 2008
Firstpage :
4121
Lastpage :
4124
Abstract :
Speaker identification systems work quite well in controlled environments but their performance degrades severely in the presence of the reverberation that is frequently encountered in realistic acoustical environments. In this paper we develop an algorithm to make speaker identification systems more robust to reverberation by passing sequences of cepstral features through a short FIR filter. The coefficients of the filter are chosen to minimize the mean square differences between compensated features in the training and testing environments. Surprisingly, the resulting filter coefficients are relatively invariant to the actual nature of the reverberation. The use of the post-filtering approach is shown to improve speaker identification accuracy, especially when reverberation times are relatively long.
Keywords :
FIR filters; cepstral analysis; distortion; mean square error methods; reverberation; speaker recognition; FIR filter; cepstral feature; environment-invariant compensation; linear post-filtering; mean square difference; minimum distortion; reverberation; speaker identification; Cepstral analysis; Control systems; Finite impulse response filter; Loudspeakers; Noise robustness; Reverberation; Speaker recognition; Speech analysis; Wiener filter; Working environment noise; Deconvolution; Least Mean Square Methods; Speaker Recognition; Wiener Filtering;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Acoustics, Speech and Signal Processing, 2008. ICASSP 2008. IEEE International Conference on
Conference_Location :
Las Vegas, NV
ISSN :
1520-6149
Print_ISBN :
978-1-4244-1483-3
Electronic_ISBN :
1520-6149
Type :
conf
DOI :
10.1109/ICASSP.2008.4518561
Filename :
4518561
Link To Document :
بازگشت