LINGUISTIC AND ACOUSTIC RESOURCES OF THE COMPUTER-BASED SYSTEM FOR ANALYSIS AND INTERPRETATION OF SPEECH INTONATION

Abstract

This article describes a novel approach to discriminating native and nonnative utterances based on suprasegmental features that constitute the intonation of the syntagma. In each language speech presents a diverse number of suprasegmental phenomena, which working together, create a certain coloring of spoken speech, intonationally giving out the emotional involvement or detachment of the speaker. To any language, there is a variety of intonational patterns, which were first identified as such by E. A. Bryzgunova, and are used for teaching Russian as a foreign language by I. V. Odintsova. According to the seven basic intonational patterns of E. A. Bryzgunova, BM Lobanov created the universal melodic portraits (UMP), which can be normalized in time in accordance with the position of its pre-nuclear, nuclear and bender parts. With the help of intonation, information is transmitted at the linguistic and paralinguistic level. A lot depends on correct intoning: perception, understanding and meaning. A thought covered in words with the right acoustic pronunciation gives a correct impression of things in human mind. Thus, the so-called interference while studying a foreign language is excluded. Linguistic and acoustic resources provide the basis for computer-aided intonation training in and outside the classroom. There is a great need for new and innovative computer-aided programs. But there is no necessary software available for such teaching system and, therefore, such system appears to be of great relevance. Presented work describes as well the algorithm of designing of the computer-based system for analysis and interpretation of speech intonation.