CONF ikbal-rr-04-19p/IDIAP Entropy Based Combination of Tandem Representations for Noise Robust ASR Ikbal, Shajith Misra, Hemant Sivadas, Sunil Hermansky, Hynek Bourlard, Hervé EXTERNAL http://publications.idiap.ch/attachments/reports/2004/rr04-19.pdf PUBLIC http://publications.idiap.ch/index.php/publications/showcite/ikbal-rr-04-19 Related documents Proceedings of the INTERSPEECH-ICSLP-04 2004 Jeju Island, Korea October 2004 To appear In this paper, we present an entropy based method to combine tandem representations of the recently proposed Phase AutoCorrelation (PAC) based features and Mel-Frequency Cepstral Coefficients (MFCC) features. PAC based features, derived from a nonlinear transformation of autocorrelation coefficients and shown to be noise robust, improve their robustness to additive noise in their tandem representation. On the other hand, MFCC features in their tandem representation show a significant improvement in recognition performance on clean speech. An entropy based combination method investigated in this paper adaptively gives a higher weighting to the representation of MFCC features in clean speech and to the representation of PAC based features in noisy speech, thus yielding a robust recognition performance in all conditions. REPORT ikbal-rr-04-19/IDIAP Entropy Based Combination of Tandem Representations for Noise Robust ASR Ikbal, Shajith Misra, Hemant Sivadas, Sunil Hermansky, Hynek Bourlard, Hervé EXTERNAL http://publications.idiap.ch/attachments/reports/2004/rr04-19.pdf PUBLIC Idiap-RR-19-2004 2004 IDIAP Martigny, Switzerland In this paper, we present an entropy based method to combine tandem representations of the recently proposed Phase AutoCorrelation (PAC) based features and Mel-Frequency Cepstral Coefficients (MFCC) features. PAC based features, derived from a nonlinear transformation of autocorrelation coefficients and shown to be noise robust, improve their robustness to additive noise in their tandem representation. On the other hand, MFCC features in their tandem representation show a significant improvement in recognition performance on clean speech. An entropy based combination method investigated in this paper adaptively gives a higher weighting to the representation of MFCC features in clean speech and to the representation of PAC based features in noisy speech, thus yielding a robust recognition performance in all conditions.