Speech errors on frequently observed homophones in French: Perceptual evaluation vs automatic classification

The present contribution aims at increasing our understanding of automatic speech recognition (ASR) errors involving frequent homophone or almost homophone words by confronting them to perceptual results. The long-term aim is to improve acoustic modelling of these items to reduce automatic transcription errors. A first question of interest is whether homophone words such as et, (and) and est (to be), for which ASR systems rely on language model weights, can be discriminated in a perceptual transcription test with similar n-gram constraints. A second question concerns the acoustic separability of the two homophone words using appropriate acoustic and prosodic attributes. The perceptual test reveals that even though automatic and perceptual errors correlate positively, human listeners in conditions attempting to approximate the information available for decision for a 4-gram language model deal with local ambiguity more efficiently than ASR systems. The corresponding acoustic analysis shows that the homophone words may be distinguished thanks to relevant acoustic and prosodic attributes. A first experiment in automatic classification of the two words using data mining techniques highlights the role of the prosodic (duration and voicing) and contextual information (co-occurrence of pauses). Preliminary results suggests that additional levels of information may be considered in order to efficiently represent and factorize the word variants observed in speech and to improve the automatic speech transcription. 1