Konstanz prosodically annotated infant-directed speech corpus (KIDS corpus)

Abstract

Knowing about infants’ input is a prerequisite for some theories of first language acquisition. Here, we present the first prosodically annotated infant-directed speech corpus in German (KIDS Corpus) – a tool for formulating hypotheses and modeling acquisition processes in the prosodic domain and at the prosody-syntax interface. The multi-layered corpus consists of 524 intonation phrases (IPs) directed to infants younger than one year (196 IPs were extracted from the CHILDES database; 328 IPs were extracted from our own recordings). Pitch accents (n=832) and boundary tones (n=1048) were labeled according to GToBI. Furthermore, we annotated the presence of unaccented syllables and pitch targets before and after the accentual syllables. We also tagged the word-prosodic structure of accented words and the syntactic category of all words. In the database, 41% of the words carry a pitch accent. Within the corpus, most words are verbs, but the words that bear a pitch accent are most often nouns. The majority of phrases start and end in low boundary tones. The most frequent pitch accent types are H* and L+H*. The data are discussed in terms of elicitation settings and potential implications of distribution frequencies on first language acquisition mechanisms.