The Information Structure–prosody interface in text-to-speech technologies. An empirical perspective

Abstract The correspondence between the communicative intention of a speaker in terms Information Structure and way this reflects aspects by means prosody have been fruitful field study Linguistics. However, text-to-speech applications still lack variability richness found human speech how humans display their communication skills. Some attempts were made past to model one aspect Structure, namely thematicity for its application intonation generation technologies. Yet, these suffer from two limitations: (i) they draw upon small number made-up simple question-answer pairs rather than on real (spoken or written) corpus material; (ii) do not explore whether any other interpretation would better suit wider range textual genres beyond dialogs. In paper, different interpretations technologies are examined: state-of-art binary (and flat) theme-rheme, hierarchical defined Igor Mel’čuk within Meaning-Text Theory. outcome experiments native speakers US English suggests that latter has versatile implementation potential – interface.