Robust speech recognition features based on temporal trajectory filtering of frequency band spectrum

The paper presents the use of a variety of filters in the temporal trajectories of the frequency band spectrum to extract speech recognition features for environmental robustness. Three kinds of filters for emphasizing the statistically important parts of speech are proposed. First, a bank of RASTA-like band-pass filters to fit the statistical peaks of the modulation frequency band spectrum of speech are used. Secondly, a three-channel octave band-filter band with a smoothed rectangular window spline is applied. Thirdly, a data-driven filter is developed. Experimental results show that significant improvements for speech recognition using the proposed feature extraction approach under noisy environments can be achieved