The Korean Corpus of Spontaneous Speech

Abstract

phonological variations that occur naturally in spontaneous speech. The interests in variability of speech are not limited to the phonology but are expanding toward other areas of linguistics and speech-related fields. Speech data recorded in a studio by reading aloud a given script are gradually being replaced more and more by those recorded spontaneously in a natural setting without any script. The two types of recorded materials are equally important in pursuing various linguistic research goals. However, it is evident that the recordings from natural spontaneous speech contain more phonetic, phonological and sociolinguistic phenomena of the native speakers of a particular language than those from unnatural citation-form studio recordings. The Buckeye corpus of spontaneous speech[1] is a pioneering example of natural speech data. The 40 hours of recording from American talkers along with a search tool[2] are available for free to the research community and the corpus has been in the front line of changing the phonetic and phonological aspects of linguistic study in the world. In the same vein, Praat[3], the free software for doing phonetics by computer, has changed the way phoneticians do their work since its introduction to the linguistic research community.

Effects of gender, age, and individual speakers on articulation rate in Seoul Korean spontaneous speech

“…The 40 speakers were obtained from a Korean corpus of spontaneous speech (Yun et al, 2015). The subjects are Seoul speakers who were born, raised, and spent most of their lives in Seoul and the Gyeonggi region.…”

Affixation effects on word-final coda deletion in spontaneous Seoul Korean speech

“…The recordings of ten native speakers of Seoul Korean were selected from the Korean Corpus of Spontaneous Speech (Yun et al, 2015). They were five men and five women who were between the ages of 20 and 29 at the time of recording.…”

“…All were raised in Seoul or the Gyeonggi region and were paid volunteers. For more details on the Korean Corpus of Spontaneous Speech, please see the corpus manual (Yun et al, 2015).…”

“…The transcribed data used in the current study were derived from the Korean Corpus of Spontaneous Speech (Yun et al, 2015). All speech data in the corpus were obtained from the interview of one sociolinguistic-form session.…”

“…There are various factors that predict deletion in the speech stream; however, the current study focuses on word positions (i.e., internal and final) and morpheme types (i.e., roots and affixes). More specifically, the current study investigates the interaction of word positions and morpheme types using a dataset drawn from the Korean Corpus of Spontaneous Speech (Yun et al, 2015). The examples of coda deletion in the reduced forms are shown below along with their corresponding citation forms for comparison.…”