Sigmoidal spectral conversion with changeable dynamic region for speech feature extraction
暂无分享,去创建一个
A noise-resistant feature extraction method for speech recognition in adverse conditions is presented. The proposed method converts the speech spectrum by adjusting the dynamic region of the spectrum based on the level dependent property of auditory nerves. With this method, it is possible to alleviate the problems caused by shrunken spectral dynamic region from heavy background noise.
[1] J. Pickles. An Introduction to the Physiology of Hearing , 1982 .
[2] Hynek Hermansky,et al. RASTA processing of speech , 1994, IEEE Trans. Speech Audio Process..
[3] John H. L. Hansen,et al. Discrete-Time Processing of Speech Signals , 1993 .
[4] Stan Davis,et al. Comparison of Parametric Representations for Monosyllabic Word Recognition in Continuously Spoken Se , 1980 .