Interactional style detection for versatile dialogue response using prosodic and semantic features

This work presents an approach to interactional style (IS) detection for versatile responses in spoken dialogue systems (SDSs). Since speakers generally express their intents in different styles, the responses of an SDS should be versatile instead of invariable, planned responses. Moreover, the IS of dialogue turns can be affected by dialogue topics and speakers' emotional states. In this study, three base-level classifiers are employed for preliminary detection, including latent Dirichlet allocation for dialogue topic categorization, support vector machine for prosody-based emotional state identification and maximum entropy for semantic label-based emotional state identification. Finally, an artificial neural network is adopted for IS detection considering the scores estimated from the aforementioned classifiers. To evaluate the proposed approach, an SDS in a chatting domain was constructed for evaluation. The performance of IS detection can achieve 82.67%.