A robust system for natural spoken dialogue

Allen, James F.; Miller, Bradford W.; Ringger, Eric K.; Sikorski, Teresa

doi:10.3115/981863.981872

Cited by 139 publications

(53 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Such an exchange might occur if there is more than one train at Boston station, and the system chose a train other than the one the user meant. Whereas the original TRAINS-96 dialogue system [3] would respond to this apparently contradictory sequence of commands by sending the very same train, our enhanced HCI system notes the contradiction, and, by assessing the problem, identifies a possible mistake in its choice of referent for 'the Boston train'. Thus, the enhanced system will choose a different train the second time around, or if there are no other trains in Boston, it will ask the user to specify the train by name.…”

Section: Mcl-enhanced Human-computer Dialoguementioning

confidence: 99%

Logic, Self-Awareness and Self-Improvement: The Metacognitive Loop and the Problem of Brittleness

Anderson¹,

Perlis²

2004

View full text Add to dashboard Cite

This essay describes a general approach to building perturbation-tolerant autonomous systems, based on the conviction that artificial agents should be able to notice when something is amiss, assess the anomaly, and guide a solution into place. This basic strategy of self-guided learning is termed the metacognitive loop; it involves the system monitoring, reasoning about, and, when necessary, altering its own decision-making components. This paper (a) argues that equipping agents with a metacognitive loop can help to overcome the brittleness problem, (b) details the metacognitive loop and its relation to our ongoing work on time-sensitive commonsense reasoning, (c) describes specific, implemented systems whose perturbation tolerance was improved by adding a metacognitive loop, and (d) outlines both short-term and long-term research agendas.

show abstract

Section: Mcl-enhanced Human-computer Dialoguementioning

confidence: 99%

Logic, Self-Awareness and Self-Improvement: The Metacognitive Loop and the Problem of Brittleness

Anderson¹,

Perlis²

2004

View full text Add to dashboard Cite

show abstract

“…The challenge arises on the basis of noisy interpretation of speech signals from the front-end ASR [2]. When a dialogue act classification module parses the user's utterance into an intention, it is known that the ASR errors might cause difficulties for the classifier, which degrades the overall performance of the entire system.…”

Section: Introductionmentioning

confidence: 99%

An empirical investigation of sparse log-linear models for improved dialogue act classification

Chen

Wang

Rudnicky

2013

2013 IEEE International Conference on Acoustics, Speech and Signal Processing

View full text Add to dashboard Cite

Previous work on dialogue act classification have primarily focused on dense generative and discriminative models. However, since the automatic speech recognition (ASR) outputs are often noisy, dense models might generate biased estimates and overfit to the training data. In this paper, we study sparse modeling approaches to improve dialogue act classification, since the sparse models maintain a compact feature space, which is robust to noise. To test this, we investigate various element-wise frequentist shrinkage models such as lasso, ridge, and elastic net, as well as structured sparsity models and a hierarchical sparsity model that embed the dependency structure and interaction among local features. In our experiments on a real-world dataset, when augmenting N -best word and phone level ASR hypotheses with confusion network features, our best sparse log-linear model obtains a relative improvement of 19.7% over a rule-based baseline, a 3.7% significant improvement over a traditional non-sparse log-linear model, and outperforms a state-of-theart SVM model by 2.2%.

show abstract

“…Spoken dialogue interfaces have been widely recognized as being highly desirable for a number of applications [1,6]. There are two important dimensions of a user interface: user-friendliness and developer-friendliness.…”

Section: Introductionmentioning

confidence: 99%

“…They are frequently system-centric in requiring rigid control of the dialogue, and frequently tied to a specific application [7], such as air travel information or train schedules [1]. Although components of these systems might be portable, there is clear need for toolkits to facilitate system construction [6].…”

Section: Introductionmentioning

confidence: 99%

Automating spoken dialogue systems

Singh

Barnett²,

Singh

1997

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Spoken dialogue interfaces apply in a number of applications. Engaging in meaningful conversation presupposes the ability to recognize and generate different conversational moves, and to adaptively carry on a dialogue. Although the portability of dialogue interfaces is highly desirable, few current approaches address it seriously. We describe a portable toolkit for constructing spoken dialogue interfaces. We present the representations and techniques used to customize an interface to a particular domain and application. Our approach relies on shallow knowledge of the domain, and interprets a rule-based model of the dialogue.

show abstract

A robust system for natural spoken dialogue

Cited by 139 publications

References 21 publications

Logic, Self-Awareness and Self-Improvement: The Metacognitive Loop and the Problem of Brittleness

Logic, Self-Awareness and Self-Improvement: The Metacognitive Loop and the Problem of Brittleness

An empirical investigation of sparse log-linear models for improved dialogue act classification

Automating spoken dialogue systems

Contact Info

Product

Resources

About