Title :
Improved Bayesian approach to robust speech segmentation
Author :
Wenjun, Zhang ; Jianying, Xie
Author_Institution :
Dept. of Autom., Shanghai Jiao Tong Univ., China
Abstract :
To enhance the robustness of speech segmentation, this paper presents the improved segmentation model based on Bayesian method which combined with a prior probability which is independent to noise feature as the compensation for the mismatch of acoustic model. We evaluated and compared the performance of different methods in speech segmentation experiment.
Keywords :
Bayes methods; hidden Markov models; speech processing; Bayesian method; acoustic model; hidden Markov models; probability; robust speech segmentation; segmentation model; Acoustic noise; Bayesian methods; Databases; Hidden Markov models; Noise robustness; Signal processing algorithms; Speech analysis; Speech enhancement; Speech recognition; Working environment noise;
Conference_Titel :
Neural Networks and Signal Processing, 2003. Proceedings of the 2003 International Conference on
Conference_Location :
Nanjing
Print_ISBN :
0-7803-7702-8
DOI :
10.1109/ICNNSP.2003.1280740
