In the present issue we discuss a theoretical model for processing of emotional speech during text analysis and synthesis. The model is developed for description of speech influence in affective mass media texts and also applies to several other types of emotional communication, including conflict, complaint and speech aggression. The model follows H-CogAff architecture of cognitive models and distinguishes units for rational inference (r-scripts) and units for emotional processing of meaning (d-scripts). We claim that certain semantic components can stimulate emotional processing, activating d-scripts and suppressing possible rational inference (r-scripts) the same semantic components appear in emotional text during text synthesis. Such affective semantic components are described as starting models of d-scripts. The model allows to describe semantic properties of emotional texts and (to certain extent) to model types of speech behaviour in emotional situations.
Cite as: Kotov, A.A. (2004) D-script model for synthesis and analysis of emotional speech. Proc. 9th Conference on Speech and Computer (SPECOM 2004), 579-585
@inproceedings{kotov04_specom,
author={Artemy A. Kotov},
title={{D-script model for synthesis and analysis of emotional speech}},
year=2004,
booktitle={Proc. 9th Conference on Speech and Computer (SPECOM 2004)},
pages={579--585}
}