Introducing the Ko Corpus of Korean Mother–Child Interaction

Abstract

We describe a corpus of speech taking place between 30 Korean mother–child pairs, divided in three groups of Prelexical (M = 0;08), Early-Lexical (M = 1;02), and Advanced-Lexical (M = 2;03). In addition to the child-directed speech (CDS), this corpus includes two different formalities of adult-directed speech (ADS), i.e., family-directed ADS (ADS_Fam) and experimenter-directed ADS (ADS_Exp). Our analysis of the MLU in CDS, family-, and experimenter-directed ADS found significant differences between CDS and ADS_Fam, and between ADS_Fam and ADS_Exp, but not between CDS and ADS_Exp. Our finding suggests that researchers should pay more attention to controlling the level of formality in CDS and ADS when comparing the two registers for their speech characteristics. The corpus was transcribed in the CHAT format of the CHILDES system, so users can easily extract data related to verbal behavior in the mother–child interaction using the CLAN program of CHILDES.

Mothers' use of touch across infants' development and its implications for word learning: Evidence from Korean dyadic interactions

“…The researchers did not communicate with the participants during the play interaction. The free-play session was followed by an adult-adult communication session, but none of the adult interactions were included in the current analyses (See Ko et al (2020) for further information about the corpus). We used the video recordings captured by the wall-mounted cameras to analyze touch events, and the audio data captured by the clip-on microphones for forced-alignment and transcription.…”

“…Data from participants in our study were collected as part of a multimodal corpus of 35 Korean mother‐child interactions, including 21 boys and 14 girls (Ko et al., 2020). Families were recruited in Seoul through online advertisements and word‐of‐mouth, and were paid for their participation.…”

“…The audio recordings of mother‐child interactions were transcribed in the CHAT format (MacWhinney, 2000) based on criteria described in Ko et al. (2020). Since one of the main goals of this study was to investigate the alignment patterns between touch and word boundaries, we needed to segment the speech into words, which we defined as a unit of morpheme(s) separated by space in transcription.…”

“…For the 3473 utterances in the corpus, we binary-coded the alignment relation with touch at the beginning and the end of each utterance. Our definition of utterance in this analysis was based on the transcription guideline as detailed in Ko et al (2020). Similar to word-touch alignment analyses, we judged an utterance and a touch to be aligned if a same-labeled edge (offset and offset, onset and onset) fell within a 500 ms window from one another (utterance-touch aligned = 1, not aligned = 0).…”

Linear Discriminative Learning: a competitive non-neural baseline for morphological inflection

“…In addition to the adult corpus, some part of a child spontaneous speech corpus, the Ko corpus (Ko et al, 2020), is provided. As with the NIKL data, the Ko corpus provides target words and their production but solely in a Romanized form.…”