During speech vocal tract is under constant influence of more than one speech segment, which causes speech movements to overlap in time and speech segments to influence one another. This phenomenon is called coarticulation and is one of the central research problems in phonetic and speech sciences. This research project is motivated by the relevance of the problem of coarticulation in the research area and by the relative lack of instrumental data on coarticulation in Croatian.

The general aim of the project is to investigate coarticulation in typical and atypical (impaired) Croatian speech. The investigation of typical speech will offer the opportunity to develop coarticulatory potential for Croatian speech sounds (the relation between coarticulatory pressure and resistance), while the investigation of atypical speech will contribute to the understanding of speech sensomotorics and its relation to coarticulation. Main contributions of the research project are: the understanding of language-specific and language-universal aspects of coarticulation as well as more successful differentiation between coarticulation in typical and atypical speech.

This project is innovative for three main reasons: 1. type of speech material (the development of communicative situations for the elicitation of quasi-natural speech in the laboratory, which will contribute to more reliable results), 2. research techniques (several instrumental techniques will be used) and 3. types of speakers (coarticulation in Croatian speech will be investigated in typical and atypical speech for the first time).

The end users are expected in the following fields: speech technology and synthesis (the development of artificial speech and human-machine communication), clinical phonetics and speech therapy (successful differentiation between typical and atypical speech as well as quantitative approach to diagnosis and therapy) and linguistics and phonetics (better understanding of human speech).