Kristiina Jokinen

From WikiPapers
Jump to: navigation, search

Kristiina Jokinen is an author.


Only those publications related to wikis are shown here.
Title Keyword(s) Published in Language DateThis property is a special property in this wiki. Abstract R C
Situated Interaction in a Multilingual Spoken Information Access Framework WikiTalk IWSDS 2014 English 18 January 2014 0 0
Evaluation of WikiTalk - User studies of human-robot interaction Evaluation
Multimodal human-robot interaction
Lecture Notes in Computer Science English 2013 The paper concerns the evaluation of Nao WikiTalk, an application that enables a Nao robot to serve as a spoken open-domain knowledge access system. With Nao WikiTalk the robot can talk about any topic the user is interested in, using Wikipedia as its knowledge source. The robot suggests some topics to start with, and the user shifts to related topics by speaking their names after the robot mentions them. The user can also switch to a totally new topic by spelling the first few letters. As well as speaking, the robot uses gestures, nods and other multimodal signals to enable clear and rich interaction. The paper describes the setup of the user studies and reports on the evaluation of the application, based on various factors reported by the 12 users who participated. The study compared the users' expectations of the robot interaction with their actual experience of the interaction. We found that the users were impressed by the lively appearance and natural gesturing of the robot, although in many respects they had higher expectations regarding the robot's presentation capabilities. However, the results are positive enough to encourage research on these lines. 0 0
Emergent verbal behaviour in human-robot interaction 2011 2nd International Conference on Cognitive Infocommunications, CogInfoCom 2011 English 2011 The paper describes emergent verbal behaviour that arises when speech components are added to a robotics simulator. In the existing simulator the robot performs its activities silently. When speech synthesis is added, the first level of emergent verbal behaviour is that the robot produces spoken monologues giving a stream of simple explanations of its movements. When speech recognition is added, human-robot interaction can be initiated by the human, using voice commands to direct the robot's movements. In addition, cooperative verbal behaviour emerges when the robot modifies its own verbal behaviour in response to being asked by the human to talk less or more. The robotics framework supports different behavioural paradigms, including finite state machines, reinforcement learning and fuzzy decisions. By combining finite state machines with the speech interface, spoken dialogue systems based on state transitions can be implemented. These dialogue systems exemplify emergent verbal behaviour that is robot-initiated: the robot asks appropriate questions in order to achieve the dialogue goal. The paper mentions current work on using Wikipedia as a knowledge base for open-domain dialogues, and suggests promising ideas for topic-tracking and robot-initiated conversational topics. 0 0