A Testbed for Examining the Timing of Feedback using a Map Task
2012 (English)In: Proceedings of the Interdisciplinary Workshop on Feedback Behaviors in Dialog, Portland, OR, 2012Conference paper (Refereed)
In this paper, we present a fully automated spoken dialogue sys-tem that can perform the Map Task with a user. By implementing a trick, the system can convincingly act as an attentive listener, without any speech recognition. An initial study is presented where we let users interact with the system and recorded the interactions. Using this data, we have then trained a Support Vector Machine on the task of identifying appropriate locations to give feedback, based on automatically extractable prosodic and contextual features. 200 ms after the end of the user’s speech, the model may identify response locations with an accuracy of 75%, as compared to a baseline of 56.3%.
Place, publisher, year, edition, pages
Portland, OR, 2012.
Computer Science Language Technology (Computational Linguistics)
IdentifiersURN: urn:nbn:se:kth:diva-107013OAI: oai:DiVA.org:kth-107013DiVA: diva2:574361
the Interdisciplinary Workshop on Feedback Behaviors in Dialog
FunderICT - The Next Generation
tmh_import_12_12_05, tmh_id_3761. QC 201301182012-12-052012-12-052013-04-11Bibliographically approved