A robust speech feature extraction procedure, by kernel regression nonlinear predictive coding, is presented. Features maximally insensitive to additive noise are obtained by growth transformation of regression functions spanning a Reproducing Kernel Hubert Space (RKHS). Experiments on TIDIGIT demonstrate consistent robustness of the new features 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 SNR.
|Journal||ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings|
|State||Published - 2004|
|Event||Proceedings - IEEE International Conference on Acoustics, Speech, and Signal Processing - Montreal, Que, Canada|
Duration: May 17 2004 → May 21 2004