Title :
Improved Two-stage Wiener Filter for Robust Speaker Identification
Author :
Xie, Yanlu ; Liu, Minghui ; Yao, Zhiqiang ; Dai, Beiqian
Author_Institution :
MOE-Microsoft Key Lab. of Multimedia Comput. & Commun., Univ. of Sci. & Technol. of China, Beijing
Abstract :
In order to solve the problem of robustness in text-independent speaker identification, two-stage Wiener filter presented in the ETSI standard is used in this paper. However, the performance of the voice activity detection (VAD) block in the standard is not so satisfying under lower SNR. Thus constrained-clustering vector quantization is presented to improve the block in the Wiener filter method. Experiments based on the MSRA´s Mandarin speech corpora show the good robustness of the improved method to several common additive noises
Keywords :
speaker recognition; vector quantisation; ETSI standard; Mandarin speech corpora; Wiener filter; constrained-clustering vector quantization; text-independent speaker identification; voice activity detection; Additive noise; Automatic speech recognition; Degradation; Noise reduction; Noise robustness; Speech enhancement; Telecommunication standards; Vector quantization; White noise; Wiener filter;
Conference_Titel :
Pattern Recognition, 2006. ICPR 2006. 18th International Conference on
Conference_Location :
Hong Kong
Print_ISBN :
0-7695-2521-0
DOI :
10.1109/ICPR.2006.696
