TY - CONF AB - Amongst several data driven approaches for designing filters for the time sequence of spectral parameters, the linear discriminant analysis (LDA) based method has been proposed for automatic speech recognition. Here we apply LDA-based filter design to cepstral features, which better match the inherent assumption of this method that feature vector components are uncorrelated. Extensive recognition experiments have been conducted both on the standard TIMIT phone recognition task and on a proprietary 130-words command word task under various adverse environmental conditions, including reverberant data with real-life room impulse responses and data processed by acoustic echo cancellation algorithms. Significant error rate reductions have been achieved when applying the novel long-range feature filters compared to standard approaches employing cepstral mean normalization and delta and delta-delta features, in particular when facing acoustic echo cancellation scenarios and room reverberation. For example, the phone accuracy on reverberated TIMIT data could be increased from 50.7\% to 56.0\% AU - Lieb, M. AU - Haeb-Umbach, Reinhold ID - 11869 KW - acoustic echo cancellation algorithms KW - adverse environmental conditions KW - automatic speech recognition KW - cepstral analysis KW - cepstral features KW - cepstral mean normalization KW - command word task KW - delta-delta features KW - delta features KW - echo suppression KW - error rate reductions KW - feature vector components KW - FIR filters KW - LDA derived cepstral trajectory filters KW - linear discriminant analysis KW - long-range feature filters KW - phone accuracy KW - real-life room impulse responses KW - reverberant data KW - spectral parameters KW - speech recognition KW - standard TIMIT phone recognition task T2 - IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2000) TI - LDA derived cepstral trajectory filters in adverse environmental conditions VL - 2 ER -