Introducción
The goal of this paper is to analyse man-machine interactions in the Spanish data of C- ORAL-ROM, a multilingual, multimedia corpus of spontaneous speech. The subcorpus consists of 41 recordings, over 2600 words. The issue we want to address is why some of the interactions succeeded, while most of them resulted in misunderstandings and failures, along with the speaker’s frustration.
The methodology is based on the analysis of the speakers’ conversational strategy, and focused on the speaker’s discursive reaction when the program does not understand the information delivered. The paper is organised in four sections. In the first one, the most relevant characteristics of the C-ORAL-ROM corpus are presented. We will focus on the man-machine conversations, and their more salient features: the transcription format, the number of words, turns, and participants. The second section will be devoted to the presentation of the human-machine dialog system used in the conversations recorded, as well as the procedure employed. The third section is dedicated to the analysis of the conversations from a pragmatic point of view. We are especially interested in the dialog strategies, the courtesy and the question-answer model. Finally, the paper provides some recommendations to improve the conversational patterns in those automatic dialog systems, to identify when the communication does not proceed properly and to design strategies to re-conduct the conversation
Add Comment