Pronunciation ambiguity vs. pronunciation variability in speech recognition

It is widely acknowledged that pronunciations in spontaneous speech di er signi cantly from citation form. For this reason, pronunciation modeling has received considerable attention in recent automatic speech recognition literature. Most of the attention however has focussed on describing an alternate pronunciation as a di erent sequence of phonetic units using the same inventory of phones which describe canonical pronunciations. Analysis of manual phonetic transcription of conversational speech reveals a large number (>20%) of cases of genuine ambiguity: instances where human labelers disagree on the identity of the surface form. In this paper, we investigate and characterize the acoustic evidence in the context of this ambiguity. We show that when a pronunciation change occurs, it is often the case that neither the canonical nor the alternate phone represent the acoustics very well. Based on this analysis, two methods for accommodating pronunciation ambiguity are developed. The rst method attempts to resolve the ambiguity by separately modeling each baseform/surfaceform pair. The second method treats the surface form as a hidden variable and \averages out" the ambiguity.