Dialogue Strategies for Improving the Usability of Telephone Human-Machine Communication

Interactions with spoken language systems may present breakdowns that are due to errors in the acoustic decoding of user utterances. Some of these errors have important consequences in reducing the naturalness of human-machine dialogues. In this paper we identify some typologies of recognition errors that cannot be recovered during the syntactico-semantic analysis, but that may be effectively approached at the dialogue level. We will describe how nonunderstanding and the effects of misrecognition are dealt with by Dialogos, a realtime spoken dialogue system that allows users to access a database of railway information by telephone. We will discuss the importance of supporting confirmation turns, and clarification and correction subdialogues. We will show the positive effects of robust dialogue management and dialogue state dependent language modeling, by taking into account both the recognition and understanding performance, and the success rate of dialogue transactions. 1 I n t r o d u c t i o n During the last few years the recognition of spontaneous speech in telephone applications "has greatly improved; nevertheless spoken dialogue between computers and inexperienced users still presents some problematic issues that reduce the user satisfaction in interacting with spoken language systems. The occurrence of errors in the acoustic decoding of users' utterance is the potential cause of miscommunication in oral interaction with spoken language systems. Some of these errors have important consequences in reducing the naturalness of humanmachine dialogues. Sometimes a robust approach in parsing and the use of language models during recognition are not sufficient to avoid recognition breakdowns. The recognition performance has a direct impact on the requirements that the dialogue modules of spoken language systems have to meet. In order to increase the usability of the applications, dialogue management modules have to deal with partial or total breakdowns of the lower levels of analysis by preventing and detecting miscommunication sources. In this paper we identify some typologies of recognition errors that cannot be recovered during the syntactico-semantic analysis, but that may be effectively approached at the dialogue level. Our analysis and the methodologies we describe have been tested in a task-oriented telephone application, but we deem that some considerations may also be useful for other display-less human-machine communication applications. We will describe how nonunderstanding and the effects of misrecognition are dealt with in Dialogos, a real-time spoken language system that allows/users to access a database of railway information by using the telephone. A detailed description of the different modules of Dialogos may be found in (Albesano et al., 1997). In this paper we will discuss the importance of supporting confirmation turns, and clarification and correction subdialogues. The dialogue module of Dialogos makes an extensive use of context knowledge: contextual information is used not only for validating or rejecting semantic interpretations, but it is also sent to the lower levels of input analysis for helping the recognizer. We will show that the positive effects of robust dialogue management and dialogue state dependent language modeling may be evaluated by taking into account both the recognition and understanding performance, and the success rate of dialogue transactions. From our experience we may conclude that if we provide robust behaviour in our dialogue systems, speech is a viable interface even with relatively low word accuracy rates. Nevertheless we believe that some important issues are still unexplored, and one of these is related to the weight that recognition errors have with respect to the degree of co-operativeness of the users. These open issues and some experimental data that emphasize their