With the aim of designing a spoken dialogue system which has the ability to adapt to the user’s communication idiosyncrasies, we investigate whether it is possible to carry over insights from the usage of communication styles in human-human interaction to human-computer interaction. In an extensive literature review, it is demonstrated that communication styles play an important role in human communication. Using a multi-lingual data set, we show that there is a significant correlation between the communication style of the system and the preceding communication style of the user. This is why two components that extend the standard architecture of spoken dialogue systems are presented: 1) a communication style classifier that automatically identifies the user communication style and 2) a communication style selection module that selects an appropriate system communication style. We consider the communication styles elaborateness and indirectness as it has been shown that they influence the user’s satisfaction and the user’s perception of a dialogue. We present a neural classification approach based on supervised learning for each task. Neural networks are trained and evaluated with features that can be automatically derived during an ongoing interaction in every spoken dialogue system. It is shown that both components yield solid results and outperform the baseline in form of a majority-class classifier.

We present a neural network approach to estimate the communication style of spoken interaction, namely the stylistic variations elaborateness and directness, and investigate which type of input features to the estimator are necessary to achive good performance. First, we describe our annotated corpus of recordings in the health care domain and analyse the corpus statistics in terms of agreement, correlation and reliability of the ratings. We use this corpus to estimate the elaborateness and the directness of each utterance. We test different feature sets consisting of dialogue act features, grammatical features and linguistic features as input for our classifier and perform classification in two and three classes. Our classifiers use only features that can be automatically derived during an ongoing interaction in any spoken dialogue system without any prior annotation. Our results show that the elaborateness can be classified by only using the dialogue act and the amount of words contained in the corresponding utterance. The directness is a more difficult classification task and additional linguistic features in form of word embeddings improve the classification results. Afterwards, we run a comparison with a support vector machine and a recurrent neural network classifier.