Automatic classification of spoken South African English variants using a transcription-less speech recognition approach

[摘要] ENGLISH ABSTRACT:We present the development of a pattern recognition system which is capable of classifyingdifferent Spoken Variants (SVs) of South African English (SAE) using a transcriptionlessspeech recognition approach. Spoken Variants (SVs) allow us to unify the linguisticconcepts of accent and dialect from a pattern recognition viewpoint. The need for the SAESV classification system arose from the multi-linguality requirement for South Africanspeech recognition applications and the costs involved in developing such applications.

[发布机构] Stellenbosch University