Paper: Constituent-based Accent Prediction

ACL ID C98-2150
Title Constituent-based Accent Prediction
Venue International Conference on Computational Linguistics
Session Main Conference
Year 1998

Near-perfect atttomatic accent assignment is at- tainable,for citation-s~le speech, but better com- putational models art, needed to predict accent in extended, spontaneous discourses. This paper presents an empirically motivated theory of the dis- course ]bcttsing nature of accent in spontaneous speech. Hypotheses based on this theory lead to a new approach to accent prediction, in which pat- terns of deviation from citation.fbrm accentuation, defined at the constituent or noun phrase level, are atttomatically learned from an annotated cor- pus. Machine learning experiments on 1031 noun phrases from eighteen spontaneous direction-giving monologues show that accent assignment c~tn be signoqcantly improved by up to 4%-6% relative to a hypothetical baseline system that would ...