Tagging Communication Problems in Spoken Dialogue Systems: On-line or Off-line?
نویسندگان
چکیده
The paper describes our work on tagging communication problems in humancomputer shared-goal spoken dialogues. The coding scheme is not otherwise restricted to a particular task or domain. The tag-set is rooted in a set of co-operativity guidelines. Tagging is done off-line at design time. The purpose is to discover problems early in the dialogue development process in order to prevent them from occurring in later versions of the dialogue. We describe the coding scheme and its application and discuss the issue of on-line versus off-line tagging of communication problems.
منابع مشابه
On-Line Learning of a Persian Spoken Dialogue System Using Real Training Data
The first spoken dialogue system developed for the Persian language is introduced. This is a ticket reservation system with Persian ASR and NLU modules. The focus of the paper is on learning the dialogue management module. In this work, real on-line training data are used during the learning process. For on-line learning, the effect of the variations of discount factor (g) on the learning speed...
متن کاملOn-Line Learning of a Persian Spoken Dialogue System Using Real Training Data
The first spoken dialogue system developed for the Persian language is introduced. This is a ticket reservation system with Persian ASR and NLU modules. The focus of the paper is on learning the dialogue management module. In this work, real on-line training data are used during the learning process. For on-line learning, the effect of the variations of discount factor (g) on the learning speed...
متن کاملCoherent Back-Channel Feedback Tagging of In-Car Spoken Dialogue Corpus
This paper describes the design of a backchannel feedback corpus and its evaluation, aiming at realizing in-car spoken dialogue systems with high responsiveness. We constructed our corpus by annotating the existing in-car spoken dialogue data with back-channel feedback timing information in an off-line environment. Our corpus can be practically used in developing dialogue systems which can prov...
متن کاملError Detection in Spoken Human-Machine Interaction
Given the state of the art of current language and speech technology, errors are unavoidable in present-day spoken dialogue systems. Therefore, one of the main concerns in dialogue design is how to decide whether or not the system has understood the user correctly. In human-human communication, dialogue participants are continuously sending and receiving signals on the status of the information...
متن کاملError Detection in Spoken Human
Given the state of the art of current language and speech technology , errors are unavoidable in present-day spoken dialogue systems. Therefore, one of the main concerns in dialogue design is how to decide whether or not the system has understood the user correctly. In human-human communication, dialogue participants are continuously sending and receiving signals on the status of the informatio...
متن کامل