The SAMMIE system

Becker, Tilman; Poller, Peter; Schehl, Jan; Blaylock, Nate; Gerstenberger, Ciprian; Kruijff-Korbayová, Ivana

doi:10.3115/1225403.1225418

Cited by 14 publications

(5 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Zentral ist dabei die situationsabhängige Auswahl einer geeigneten Kommunikationsstrategie. In [25] und [26] wird beispielsweise anhand von Wizard-of-Oz-Experimenten belegt, dass Menschen bei der zweckrationalen Interaktion die Art und Weise, wie sie mit dem Dialogpartner kommunizieren, an die Mög-lichkeiten der zur Verfügung stehenden Modalitäten, aber auch an die vom Dialogpartner gewählte Formulierung anpassen. Diese Aufgabe ist also orthogonal zum Management der Kooperation; betrifft sie doch vor allem die tatsächliche Umsetzung von Kommunikationsvorgängen in eine Reihe von Sprechakten.…”

Section: Kollaboration Und Interaktion Als Eigenständige Domänenunclassified

Task-Analysen in der Mensch-Maschine-Interaktion

Ludwig

2015

Xpert.press

View full text Add to dashboard Cite

show abstract

Section: Kollaboration Und Interaktion Als Eigenständige Domänenunclassified

Task-Analysen in der Mensch-Maschine-Interaktion

Ludwig

2015

Xpert.press

View full text Add to dashboard Cite

show abstract

“…The user is not constrained to follow this ordering, and can also over-provide information using mixed-initiative behaviour, such as over-answering. Related work replaces the default ordering of slots by a cluster-based method which determines the next most informative slot to ask at each point in the dialogue (Becker et al 2007).…”

Section: Experimental Frameworkmentioning

confidence: 99%

Does this list contain what you were searching for? Learning adaptive dialogue strategies for interactive question answering

Rieser¹,

Lemon²

2009

Nat. Lang. Eng.

View full text Add to dashboard Cite

Policy learning is an active topic in dialogue systems research, but it has not been explored in relation to interactive question answering (IQA). We take a first step in learning adaptive interaction policies for question answering : we address the question of how to acquire enough reliable query constraints, how many database results to present to the user and when to present them, given the competing trade-offs between the length of the answer list, the length of the interaction, the type of database and the noise in the communication channel. The operating conditions are reflected in an objective function which we use to derive a hand-coded threshold-based policy and rewards to train a reinforcement learning policy. The same objective function is used for evaluation. We show that we can learn strategies for this complex trade-off problem which perform significantly better than a variety of hand-coded policies, for a wide range of noise conditions, user types, types of DB and turn-penalties. Our policy learning framework thus covers a wide spectrum of operating conditions. The learned policies produce an averagerelativeincrease in reward of 86.78% over the hand-coded policies. In 93% of the cases the learned policies perform significantly better than the hand-coded ones (p< .001). Furthermore we show that the type of database has a significant effect on learning and we give qualitative descriptions of the learned IQA policies.

show abstract

“…Within the TALK project, an in-car spoken dialogue system for the MP3 domain was created and integrated into a BMW 3-Series Coupe (Becker et al, 2007). For the speech understanding component, a German corpus named SAMMIE (SAarbrücken Multi-Modal Interface Experiment) was collected by Saarland University and DFKI 2 using a Wizard of Oz experiment.…”

Section: The Nuance Sammie Grammarmentioning

confidence: 99%

“…For a more detailed description of the grammar, see (Becker et al, 2007). The German Sammie grammar in Nuance format (NuanceGer) was checked and extended continuously while the dialogue system was built.…”

Section: The Nuance Sammie Grammarmentioning

confidence: 99%

Dialogue system localization with the GF resource grammar library

Perera

Ranta

2007

Proceedings of the Workshop on Grammar-Based Approaches to Spoken Language Processing - SLP '07

View full text Add to dashboard Cite

We present two experiments in the localization of spoken dialogue systems. The domain of the dialogue system is an MP3 application for automobiles. In the first experiment, a grammar in Nuance GSL format was rewritten in Grammatical Framework (GF). Within GF, the grammar was extended from two to six languages, giving a baseline for semantically complete grammars. In the second experiment, the German version of this baseline GF grammar was extended with the goal to restore the coverage of the original Nuance grammar.

show abstract

The SAMMIE system

Cited by 14 publications

References 3 publications

Task-Analysen in der Mensch-Maschine-Interaktion

Task-Analysen in der Mensch-Maschine-Interaktion

Does this list contain what you were searching for? Learning adaptive dialogue strategies for interactive question answering

Dialogue system localization with the GF resource grammar library

Contact Info

Product

Resources

About