Phonetic annotation in spoken language corpora: a methodological overview

Abstract

The paper discusses different ways of capturing detailed phonetic information when transcribing spoken corpus data. It weighs the pros and cons of two approaches: transcription (i.e. a phone-by-phone representation) and annotation (i.e. providing tags for various well-defined phonetic phenomena of interest at the level of corpus positions).

Keywords

• phonetics
• transcription
• spoken corpora
• methodology