Discovering Regularities in Non-Native Speech

This paper presents ongoing collaborative research which focuses on the application of computational linguistic techniques to the analysis of a corpus of native and non-native speech. The aim of this research is to use computational tools for phonological acquisition and representation to identify regularities and sub-regularities between different speaker groups. The corpus is being collected and annotated at different levels as part of ongoing research into the acquisition of prosody by non-native speakers at the University of Bielefeld (see Milde & Gut, 2001). The computational tools have been designed and implemented at University College Dublin as part of a suite of tools aimed at providing a development environment for modeling, testing and evaluating phonotactic descriptions of lesser studies languages (Carson-Berndsen, 2002). The two hitherto separate research directions have now come together to apply computational linguistic tools to a corpus-based investigation of non-native speaker phonotactics. The term phonotactics refers to the permissible combination of sounds in a language. There are various ways of acquiring representations of phonotactic constraints. One approach is to manually construct a set of rules based on the linguistic intuitions of a native speaker. Another approach is to learn such constraints from a data set. The latter is the approach taken in this paper.