A Data-driven Model for Timing Feedback in a Map Task Dialogue System
2013 (English)In: 14th Annual Meeting of the Special Interest Group on Discourse and Dialogue - SIGdial, Metz, France, 2013, 375-383 p.Conference paper (Refereed)
We present a data-driven model for detecting suitable response locations in the user’s speech. The model has been trained on human–machine dialogue data and implemented and tested in a spoken dialogue system that can perform the Map Task with users. To our knowledge, this is the first example of a dialogue system that uses automatically extracted syntactic, prosodic and contextual features for online detection of response locations. A subjective evaluation of the dialogue system suggests that interactions with a system using our trained model were perceived significantly better than those with a system using a model that made decisions at random.
Place, publisher, year, edition, pages
Metz, France, 2013. 375-383 p.
Computer Science Language Technology (Computational Linguistics)
IdentifiersURN: urn:nbn:se:kth:diva-134935OAI: oai:DiVA.org:kth-134935DiVA: diva2:668770
14th Annual Meeting of the Special Interest Group on Discourse and Dialogue - SIGdial
tmh_import_13_12_02, tmh_id_3863. QC 201403202013-12-022013-12-022014-03-20Bibliographically approved