View Full Document

7 views

Unformatted text preview:

1842 IEEE TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING VOL 15 NO 6 AUGUST 2007 Robust Speech Feature Extraction by Growth Transformation in Reproducing Kernel Hilbert Space Shantanu Chakrabartty Member IEEE Yunbin Deng Member IEEE and Gert Cauwenberghs Senior Member IEEE Abstract The performance of speech recognition systems depends on consistent quality of the speech features across variable environmental conditions encountered during training and evaluation This paper presents a kernel based nonlinear predictive coding procedure that yields speech features which are robust to nonstationary noise contaminating the speech signal Features maximally insensitive to additive noise are obtained by growth transformation of regression functions that span a reproducing kernel Hilbert space RKHS The features are normalized by construction and extract information pertaining to higher order statistical correlations in the speech signal Experiments with the TI DIGIT database demonstrate consistent robustness to noise of varying statistics yielding significant improvements in digit recognition accuracy over identical models trained using Mel scale cepstral features and evaluated at noise levels between 0 and 30 dB signal to noise ratio Index Terms Feature extraction growth transforms noise robustness nonlinear signal processing reproducing kernel Hilbert Space speaker verification I INTRODUCTION HILE most current speech recognizers give acceptable recognition accuracy for clean speech their performance degrades when subjected to noise present in practical environments 1 For instance it has been observed that additive white noise severely degrades the performance of Mel cepstra based recognition systems 1 2 This performance degradation has been attributed to unavoidable mismatch between training and recognition conditions Therefore in literature several approaches have been presented for alleviating the effects of mismatch These methods can be broadly categorized as



Access the best Study Guides, Lecture Notes and Practice Exams

Loading Unlocking...
Login

Join to view Robust Speech Feature Extraction and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Robust Speech Feature Extraction and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?