Consonant discrimination in elicited and spontaneous speech: a case for signal-adaptive front ends in ASR

The constant frame length in typical ASR front ends is too long to capture transient phenomena in speech, such as stop bursts. How- ever, current HMM systems have consistently outperformed sys- tems based solely on non-uniform units. This work investigates an approach to "add back" such transient information to a speech recognizer, without losing the robustness of the standard a coustic models. We demonstrate a set of phonetically-motivated acoustic features that discriminate a preliminary test set of highly ambigu- ous voiceless stops in CV contexts. The features are automatically computed from data that had been hand-marked for consonant burst location and voicing onset (extension to automatic marking is also proposed). Two corpora are processed using a parallel set of fea- tures: conversational speech over the telephone (Switchboard), and a corpus of carefully elicited speech. The latter provides a n upper bound on discrimination, and allows for comparison of feature us- age across speaking style. We explore data-driven approaches to obtaining variable-length time-localized features compatible with an HMM statistical framework. We also suggest techniques for ex- tension to automatic annotation of burst location, for computation of features at such points, and for augmentation of an HMM system with the added information.