MULTIMODAL POETIC CORPUS: PRINCIPLES OF CONSTRUCTION AND MARKUP FEATURES

The article presents a new type of corpus within the Multimodal module of the RNC, which contains not only transcripts of audio texts, but also audio/video recordings of their pronunciation. The multimodal module presently includes the main Multimodal Corpus (MURCO) for the study of oral texts of diff erent genres, and two Multimodal Parallel Corpora (MultiPARС), which allow the comparative study of the same utterance spoken by diff erent speakers in the same or similar circumstances. Russian MultiPARС — in diff erent productions and adaptations of the same play in Russian, English-Russian MultiPARС — in productions of plays in Russian and in English, or in the original and dubbed versions of fi lms. The Multimodal Poetry Corpus is designed to explore the features of a sounding poetic text. It combines the features of multimedia, parallel and poetry corpus. It contains the poetic texts in author’s, professional (actor’s) and non-professional performance. The composition and structure of the corpus and the prospects for its use are discussed. The corpus allows to compare the written text with its sound, to compare diff erent interpretations of the sounding verse, which is of interest for the study of phonetic and pro- sodic phenomena. It also opens up many possibilities for verse studies, research of poetic syntax and semantics, etc.