We explore the utility of individually selected modulation spectral features for speech and speaker characterization in general, and specifically to prediction of the perceived speaker personality profile. We suggest a method of construction of a sparse feature space and a method of finding the approximately best feature subset for attributing a specific characteristic of speech or speaker. The current selection method is based on the Kolmogorov-Smirnov statistical test applied to individual features. We assume that the characterization task is defined empirically and no a-priory theory exist to explain characteristic attribution processes. Experimental results indicate that employment of selected modulation spectral features works better than the current state-of-the-art in prediction of personality traits.

Index Terms: speech characterization, modulation spectrum analysis, feature selection