Abstract
A new approach for intonation stylization that enables the extraction of an intonation representation from prosodically unlabeled data is introduced. This approach yields global and local intonation contour classes arising from a contour-based, parametric and superpositional intonation stylization. Based on findings about the linguistic interpretation of the contour classes derived from corpus statistics and perception experiments, we created simple prediction models for the partial generation of intonation contours from discourse structure defined by discourse segment boundaries and the information status of nouns within these segments. The predicted intonation contours were evaluated by human judgments of adequacy that yielded a high accordance.
Item Type: | Journal article |
---|---|
Form of publication: | Publisher's Version |
Keywords: | Computational intonation stylization; Data-driven; Contour-based; Superposition; Discourse structure |
Faculties: | Languages and Literatures > Department 2 > Speech Science |
Subjects: | 400 Language > 410 Linguistics |
Language: | English |
Item ID: | 22776 |
Date Deposited: | 09. Feb 2015, 07:46 |
Last Modified: | 04. Nov 2020, 13:03 |