Phonotactic and discourse aspects of content design in AESOP (Asian English Speech cOrpus Project)

Abstract

This paper reports an on-going multinational collaboration “Asian English Speech cOrpus Project” (AESOP). The aim of the consortium is to build up an open resource English speech corpus representing the varieties of English spoken in Asia. It was agreed in 2008 that each participating member will use the same recording platform to collect a common set of core data so that a set of core properties common to all varieties of Asian English, as well as to features that are particular to individual varieties can be developed. Areas of overlap between L1 and L2 English and differences among varieties of Asian English could be determined. Initial efforts in 2009 were devoted to designing spoken-language tasks that aims to elicit the phonetic aspect of a large range of English segmental and suprasegmental characteristics. The 2011 efforts, reported here by the Taiwan members, were devoted to designing spoken language tasks that aims to elicit (1) a comprehensive coverage of phonotactic variations, (2) broad vs. narrow focus and (3) dialogue and discourse materials using discourse completion tasks (DCT). We believe the data collected can be used to develop ASR tools, investigate various issues of speech prosody, and assess English proficiency towards CALL applications.