Push-to-talk ain’t always bad! Comparing Different Interactivity Settings in Task-oriented Dialogue
نویسندگان
چکیده
Restrictions of interactivity in dialogue are often seen as having negative impact on the efficiency of the dialogue, as they affect the ability to give immediate feedback (Whittaker, 2003). We have conducted experiments with one such restriction common in spoken dialogue systems, namely push-totalk. While our results confirm many predictions from the literature (fewer but longer turns; reduction of positive feedback), we found no significant impact on task-efficiency. Our analysis of the grounding strategies of the subjects shows that the restriction actually induced a more cautious strategy that proved advantageous for our matching task, and that giving negative feedback in the form of clarification requests was not affected by the restriction.
منابع مشابه
Referring under Restricted Interactivity Conditions
We report results on how the collaborative process of referring in task-oriented dialogue is affected by the restrictive interactivity of a turn-taking policy commonly used in dialogue systems, namely push-to-talk. Our findings show that the restriction did not have a negative effect. Instead, the stricter control imposed at the interaction level favoured longer, more effective referring expres...
متن کاملPentoRef: A Corpus of Spoken References in Task-oriented Dialogues
PentoRef is a corpus of task-oriented dialogues collected in systematically manipulated settings. The corpus is multilingual, with English and German sections, and overall comprises more than 20000 utterances. The dialogues are fully transcribed and annotated with referring expressions mapped to objects in corresponding visual scenes, which makes the corpus a rich resource for research on spoke...
متن کاملThe Influence of Scenario Constraints on the Spontaneity of Speech. A Comparison of Dialogue Corpora
In this article we compare two large scale dialogue corpora recorded in different settings. The main differences are unrestricted turntaking vs. push-to-talk button and complex vs. simple negotiation task. In our investigation we found that vocabulary, durations of turns, words and sounds as well as prosodical features are influenced by differences in the setting.
متن کاملEnd-to-End Optimization of Task-Oriented Dialogue Model with Deep Reinforcement Learning
In this paper, we present a neural network based task-oriented dialogue system that can be optimized end-to-end with deep reinforcement learning (RL). The system is able to track dialogue state, interface with knowledge bases, and incorporate query results into agent’s responses to successfully complete task-oriented dialogues. dialogue policy learning is conducted with a hybrid supervised and ...
متن کاملBeyond Repair – Testing the Limits of the Conversational Repair System
We report on an experiment on the effects of inducing acoustic understanding problems in task-oriented dialogue. We found that despite causing real problems w.r.t. task performance, many instances of induced problems were not explicitly repaired by the dialogue participants. Almost all repairs referred to the immediately preceding utterance, with problems in prior utterances left unacknowledged...
متن کامل