State Tracking & Dialog Systems 2015


Notes:

In dialog systems, “state tracking” – sometimes also called “belief tracking” – refers to accurately estimating the user’s goal as a dialog progresses.

  • Dialog state
  • Dialog state tracker
  • Dialog state trackers
  • Rule based tracker
  • Stacked relational trees

Resources:

See also:

Conditional Random Fields & Dialog Systems 2013 | Conditional Random Fields & Dialog Systems 2014 | Deep Learning & Dialog Systems 2014 | Dialog Systems 2013 | Parse Tree & Dialog Systems 2013 | Question-Answer Pairs & Dialog Systems | RNN (Recurrent Neural Network) & Dialog Systems 2014 | SVM (Support Vector Machine) & Dialog Systems 2013 | SVM (Support Vector Machine) & Dialog Systems 2014


Multi-domain dialog state tracking using recurrent neural networks N Mrkši?, DO Séaghdha, B Thomson, M Gaši?… – arXiv preprint arXiv: …, 2015 – arxiv.org … There has been a small amount of previous work on domain adapta- tion for dialog systems. Tur et al. … To the best of our knowledge, our work is the first attempt to build a belief tracker capable of operating across disjoint dialog domains. 3 Dialog State Tracking using RNNs … Cited by 14 Related articles All 16 versions

The ubuntu dialogue corpus: A large dataset for research in unstructured multi-turn dialogue systems R Lowe, N Pow, I Serban, J Pineau – arXiv preprint arXiv:1506.08909, 2015 – arxiv.org … Similar progress has not yet been observed in the development of dialogue systems. … The dataset is orders of magnitude larger than struc- tured corpuses such as those of the Dialogue State Tracking Challenge [30]. It is on the same scale as recent datasets for solving problems … Cited by 20 Related articles All 12 versions

Learning from real users: Rating dialogue success with neural networks for reinforcement learning in spoken dialogue systems PH Su, D Vandyke, M Gasic, D Kim, N Mrksic… – arXiv preprint arXiv: …, 2015 – arxiv.org … Index Terms: spoken dialogue systems, real users, reward pre- diction, dialogue success classification, neural network … ASR, a confusion net- work (CNet) semantic input decoder [28], the BUDS [19] belief state tracker that factorises the dialogue state using a dynamic … Cited by 10 Related articles All 12 versions

Incremental LSTM-based dialog state tracker L Zilka, F Jurcicek – 2015 IEEE Workshop on Automatic Speech …, 2015 – ieeexplore.ieee.org … For example, in the restaurant information domain, the dialog state tracker can track what kind of food the user … First, they can only track the dialog state turn- by-turn (as opposed to a more … which limits their interaction with users: For ex- ample, in a typical dialog system [7, 8], the … Cited by 6 Related articles All 4 versions

Knowledge graph inference for spoken dialog systems Y Ma, PA Crook, R Sarikaya… – 2015 IEEE International …, 2015 – ieeexplore.ieee.org … into Markov Random Fields in order to create user goal track- ing models that could form part of a spoken dialog system. Since semantic knowledge graphs include both entities and their attributes, the proposed method merges the semantic dialog-state-tracking of attributes and … Cited by 6 Related articles All 6 versions

Users’ Belief Awareness in Reinforcement Learning-Based Situated Human–Robot Dialogue Management E Ferreira, G Milliez, F Lefèvre, R Alami – … Language Dialog Systems and …, 2015 – Springer … In this paper, we described how a user belief real-time tracking framework can be used along with a multimodal POMDP-based dialogue management. … Thomson B, Young S (2010) Bayesian update of dialogue state: a pomdp framework for spoken dialogue systems. … Cited by 9 Related articles All 7 versions

A neural network approach to context-sensitive generation of conversational responses A Sordoni, M Galley, M Auli, C Brockett, Y Ji… – arXiv preprint arXiv: …, 2015 – arxiv.org … While there are previous uses of ma- chine learning for response generation (Walker et al., 2003), dialog state tracking (Young et al., 2010), and user modeling (Georgila et al., 2006), many compo- nents of typical dialog systems remain hand-coded: in particular, the labels and … Cited by 48 Related articles All 15 versions

A survey of available corpora for building data-driven dialogue systems IV Serban, R Lowe, L Charlin, J Pineau – arXiv preprint arXiv:1512.05742, 2015 – arxiv.org … implies, these datasets are used to learn a strategy for the Dialogue State Tracker (sometimes called … State tracking is useful as it can increase the robustness of speech recognition systems, and can provide an im- plementable framework for real-world dialogue systems. … Cited by 8 Related articles All 2 versions

Constrained markov bayesian polynomial for efficient dialogue state tracking K Yu, K Sun, L Chen, S Zhu – IEEE/ACM Transactions on Audio, …, 2015 – ieeexplore.ieee.org … contrast, discriminative state tracking models have been successfully used for spoken dialogue systems [7]. The … In DSTC-3, where the task is to adapt a state tracker to a … data-driven framework, there have been attempts to employ rule-based methods for dialogue state tracking. … Cited by 4 Related articles All 4 versions

Reinforcement-learning based dialogue system for human–robot interactions with socially-inspired rewards E Ferreira, F Lefevre – Computer Speech & Language, 2015 – Elsevier … (1998)) has been successfully employed in the Spoken Dialogue System (SDS) field (Young et al., 2010, Thomson and Young, 2010 … In all non-stationary contexts (eg environment, optimization method) tracking the value function instead of converging to it seems preferable. … Cited by 8 Related articles All 3 versions

Evaluating prerequisite qualities for learning end-to-end dialog systems J Dodge, A Gane, X Zhang, A Bordes, S Chopra… – arXiv preprint arXiv: …, 2015 – arxiv.org … & Le, 2015; Sordoni et al., 2015) directly generate a response given the last user utterance and (potentially) the context from previous dialog turns without relying on the intermediate use of a dialog state tracking component like in traditional dialog systems (eg in Henderson … Cited by 10 Related articles All 3 versions

A hybrid approach to dialogue management based on probabilistic rules P Lison – Computer Speech & Language, 2015 – Elsevier … of dialogue strategies is a challenging task in the development of spoken dialogue systems (SDS … The selection of system actions is often grounded in a complex dialogue state encompassing a … The robot’s tracking of the current dialogue state is therefore bound to remain partial … Cited by 16 Related articles All 7 versions

Stochastic language generation in dialogue using recurrent neural networks with convolutional sentence reranking TH Wen, M Gasic, D Kim, N Mrksic, PH Su… – arXiv preprint arXiv: …, 2015 – arxiv.org … The latter is par- ticularly important in spoken dialogue systems where frequent repetition of identical output forms t. The work reported in this paper is part of a larger … 2014. Robust dialog state tracking using delexicalised recurrent neural networks and unsu- pervised adaptation. … Cited by 16 Related articles All 19 versions

Machine learning for dialog state tracking: A review M Henderson – 2015 – research.google.com … Young, “Bayesian update of dialogue state: A POMDP framework for spoken dialogue systems,” Com- puter … J. Lee, W. Lim, and K.-E. Kim, “Optimizing generative dialog state tracker via cascading … E. Kim, J. Lee, and J. Sohn, “Engineer- ing statistical dialog state trackers: A case … Cited by 3 Related articles All 2 versions

Hierarchical neural network generative models for movie dialogues IV Serban, A Sordoni, Y Bengio, A Courville… – arXiv preprint arXiv: …, 2015 – arxiv.org … Dialogue systems, also known as interactive con- versational agents, virtual agents and sometimes chatterbots, are used in a wide set of applications ranging from technical support services to lan- guage learning tools and entertainment (Young et al., 2013; Shawar and Atwell … Cited by 16 Related articles All 4 versions

Recurrent polynomial network for dialogue state tracking K Sun, Q Xie, K Yu – arXiv preprint arXiv:1507.03934, 2015 – arxiv.org … Figure 1: Diagram of a spoken dialogue system (SDS) … 4. RPN for Dialogue State Tracking As introduced in section 1, in this paper, the dialogue state tracker receives SLU N-best hypotheses for each user turn, each hypothesis having a set of act-slot-value tuples with a … Cited by 4 Related articles All 5 versions

Scalable summary-state pomdp hybrid dialog system for multiple goal drifting requests and massive slot entity instances S Koo, S Ryu, K Lee, GG Lee – … Dialog Systems and Intelligent Assistants, 2015 – Springer … Previous architectures required that SDS designers build new a tracker model and that they assign new reward values … tics, Stroudsburg, PA, pp 149–152 Thomson B, Young S (2010) Bayesian update of dialogue state: a pomdp framework for spoken dialogue systems. … Cited by 2 Related articles All 5 versions

Discriminative methods for statistical spoken dialogue systems MS Henderson – 2015 – repository.cam.ac.uk … Discriminative DST using RNNs, including word-based trackers (which do not use any explicit SLU), are also evalu- … Automatic Speech Recognition Dialogue State Tracking Dialogue Management … user Figure 2.1: The pipeline architecture of a typical dialogue system. 5 … Cited by 4 Related articles All 3 versions

From rule-based to data-driven lexical entrainment models in spoken dialog systems J Lopes, M Eskenazi, I Trancoso – Computer Speech & Language, 2015 – Elsevier … 1. Introduction. The use of Spoken Dialog Systems (SDSs) in everyday life is getting closer to being a reality. However … tests. In this version, same dialog-state tracking module used in Lets Go replaced the agenda-based dialog manager. … Cited by 4 Related articles All 5 versions

Online adaptative zero-shot learning spoken language understanding using word-embedding E Ferreira, B Jabaian, F Lefèvre – 2015 IEEE International …, 2015 – ieeexplore.ieee.org … understanding module on the second Dialog State Tracking Challenge (DSTC2) datasets. Index Terms— Spoken language understanding, word embedding, zero-shot learning, out-of-domain training data, online adaptation. 1. INTRODUCTION In dialogue systems, the Spoken … Cited by 5 Related articles All 2 versions

Reward shaping with recurrent neural networks for speeding up on-line policy learning in spoken dialogue systems PH Su, D Vandyke, M Gasic, N Mrksic, TH Wen… – arXiv preprint arXiv: …, 2015 – arxiv.org … Spoken dialogue systems (SDS) offer a natural way for people to interact with computers. … a confusion network (CNet) semantic input decoder (Henderson et al., 2012), the BUDS (Thomson and Young, 2010) belief state tracker that factorises the dialogue state using a … Cited by 2 Related articles All 18 versions

Hypotheses Ranking and State Tracking for a Multi-Domain Dialog System Using Multiple ASR Alternates OZ Khan, JP Robichaud, P Crook… – … Conference of the …, 2015 – pdfs.semanticscholar.org … additional results from ASR. Index Terms: dialog systems, natural language understanding, speech recognition, hypotheses ranking, dialog state tracking, multi-domain classification, contextual domain classification 1. Introduction … Cited by 2 Related articles All 3 versions

LecTrack: Incremental Dialog State Tracking with Long Short-Term Memory Networks L Žilka, F Jur?í?ek – International Conference on Text, Speech, and …, 2015 – Springer … Third, elaborate and complicated tracking models of the trackers are difficult to reproduce and … The contribution of this paper is our novel dialog state tracker, which we refer to as … 1. This paper aims towards building more responsive and simpler dialog systems by proposing the … Cited by 1 Related articles

Incremental coordination: Attention-centric speech production in a physically situated conversational agent Z Yu, D Bohus, E Horvitz – 16th Annual Meeting of the Special …, 2015 – anthology.aclweb.org … to coordinate speech production with the listeners’ focus of attention in a physically situated dialog system. … The model leverages features from visual subsystems (eg, face detection and tracking, head-pose … via a heuristic rule that takes into ac- count the dialog state and the … Cited by 6 Related articles All 14 versions

Conversational system for information navigation based on POMDP with user focus tracking K Yoshino, T Kawahara – Computer Speech & Language, 2015 – Elsevier … Abstract. We address a spoken dialogue system which conducts information navigation in a style of small talk. … In this work, we formulate the problem of dialogue management as a selection of modules and optimize it with POMDP by tracking the dialogue state and focus of … Cited by 3 Related articles All 3 versions

Implementation of generic positive-negative tracker in extensible dialog system S Koo, S Ryu, GG Lee – 2015 IEEE Workshop on Automatic …, 2015 – ieeexplore.ieee.org … slot entity instances,” in 2015 International Workshop Series on Spoken Dialogue Systems Technology, 2015 … Comparison of bayesian discrim- inative and generative models for dialogue state tracking,” in Proceedings … and Kai Yu, “A gen- eralized rule based tracker for dialogue … Cited by 1 Related articles

Multi-domain dialogue success classifiers for policy training D Vandyke, PH Su, M Gasic, N Mrksic… – … IEEE Workshop on …, 2015 – ieeexplore.ieee.org … most likely user action (as estimated by the belief tracker) ? full belief state … David Vandyke, Tsung-Hsien Wen, and Steve Young, “Multi-domain dialog state tracking using recurrent … and Shimei Pan, “Designing and eval- uating an adaptive spoken dialogue system,” User Mod … Cited by 3 Related articles All 8 versions

Knowledge transfer between speakers for personalised dialogue management I Casanueva, T Hain, H Christensen… – … Annual Meeting of …, 2015 – anthology.aclweb.org … For each speaker, the state tracker has been trained with data from the other 14 … been tested 4Because of the variable depth tree structure of the spoken dialogue system, the sum … for speaker specific dialogue management and could be used in other tasks such as state tracking. … Cited by 4 Related articles All 10 versions

Policy committee for adaptation in multi-domain spoken dialogue systems M Gaši?, N Mrkši?, PH Su, D Vandyke, TH Wen… – 2015 – repository.cam.ac.uk … work we assume that the spoken language understand- ing unit, the belief tracker and the … [17] B Thomson and S Young, “Bayesian update of dialogue state: A POMDP … SJ Young, “Agenda-Based User Simulation for Boot- strapping a POMDP Dialogue System,” in Proceedings … Cited by 4 Related articles All 7 versions

Recurrent Polynomial Network for Dialogue State Tracking with Mismatched Semantic Parsers Q Xie, K Sun, S Zhu, L Chen… – 16th Annual Meeting of …, 2015 – anthology.aclweb.org … in figure 2. 4 Uncertainty in SLU In an end-to-end dialogue system, there are two … because the confidence scores of SLU hypotheses are usually the key in- puts for dialogue state tracking. When these confi- dence scores become unreliable, the performance of tracker is sure to … Cited by 1 Related articles All 10 versions

Dialog State Tracking Challenge 4 S Kim, LF D’Haro, RE Banchs, J Williams… – 2015 – colips.org … this fourth edition of the Dialog State Tracking Challenge, we will focus on a dialog state tracking task on … a series of pilot tracks for the core components in developing end-to-end dialog systems based on … 1, 2 from tour guide-2 and 2 from tour guide-3) for evaluating the trackers. … Cited by 1 Related articles All 2 versions

Interaction quality: assessing the quality of ongoing spoken dialog interaction by experts—and how it relates to user satisfaction A Schmitt, S Ultes – Speech Communication, 2015 – Elsevier … are expected to deliver a more stable rating when evaluating the performance of a dialog system. … similar and recent data-driven approach addresses the recognition of the dialog state ( Williams et … This would imply that tracking user satisfaction is user-specific and would require … Cited by 6 Related articles All 4 versions

Yarbus: Yet another rule based belief update system J Fix, H Frezza-Buet – arXiv preprint arXiv:1507.06837, 2015 – arxiv.org … The recently proposed dialog state tracking challenges offer the opportunity to test belief tracking algorithms on … The tracker proposed by Williams (2014) ranked first at the time of the … solve the belief tracking problem without requiring much of expert knowledge in dialog systems. … Cited by 2 Related articles All 3 versions

The Influence of Context on Dialogue Act Recognition E Ribeiro, R Ribeiro, DM de Matos – arXiv preprint arXiv:1506.00839, 2015 – arxiv.org … The Cambridge Restaurant Corpus was the corpus used in the Second Dia- logue State Tracking Challenge [21]. It features data extracted from telephonic … However, this is not useful for a dialogue system trying to identify the intention of its conversational … Cited by 3 Related articles All 4 versions

Using knowledge on word-islands to improve the performance of spoken dialogue systems R López-Cózar – Knowledge-Based Systems, 2015 – Elsevier … address this second drawback by inverting the conventional notion of dialogue state, and are … This approach combines the benefits of belief state tracking and RL, thereby providing a … a different approach to improve the performance of spoken dialogue systems, the technique … Cited by 4 Related articles All 3 versions

Deep Contextual Language Understanding in Spoken Dialogue Systems C Liu, P Xu, R Sarikaya – Sixteenth Annual Conference of …, 2015 – research.microsoft.com … time entities in the second turn. In many spoken dialog systems, DM plays the central role in resolving such contextual ambiguity. It is usually formulated as the task of dialog state tracking [5] in the literature. The goal of the task … Cited by 2 Related articles All 4 versions

Fix it where it fails: Pronunciation learning by mining error corrections from speech logs Z Kou, D Stanton, F Peng, F Beaufays… – … on Acoustics, Speech …, 2015 – ieeexplore.ieee.org … Orlandi, Christopher Culy, and Horacio Franco, “Using dialog corrections to improve speech recognition,” in Error Handling in Spoken Language Dialogue Systems, 2003. … [12] Jason D. Williams, “Exploiting the asr n-best by tracking mul- tiple dialog state hypotheses,” in … Cited by 5 Related articles All 6 versions

Human-machine dialogue as a stochastic game M Barlier, J Perolat, R Laroche, O Pietquin – 16th Annual SIGdial Meeting …, 2015 – hal.inria.fr … In a Spoken Dialogue System (SDS), the Dia- logue Manager (DM) is designed in order to im … has to take a decision about what to say next according to the dialogue context (also called dialogue state). … Equilibrium selection and tracking may be a big deal while working with SGs … Cited by 4 Related articles All 16 versions

Learning Domain-Independent Dialogue Policies via Ontology Parameterisation Z Wang, TH Wen, PH Su, Y Stylianou – … of the Special Interest Group on …, 2015 – aclweb.org … 2014. A gener- alized rule based tracker for dialogue state tracking. In SLT 2014. Blaise Thomson and Steve Young. 2010. Bayesian update of dialogue state: A POMDP framework for spoken dialogue systems. Computer Speech and Language, 24 (4): 562–588. … Cited by 2 Related articles All 16 versions

Multi-Language Hypotheses Ranking And Domain Tracking for Open Domain Dialogue Systems PA Crook, JP Robichaud… – … Annual Conference of …, 2015 – research.microsoft.com … K. Sagae, PS Georgiou, DR Traum, and SS Narayanan, “A reranking approach for recognition and classification of speech input in conversational dialogue systems.” in IEEE … [7] JD Williams, “Web-style ranking and slu combination for dialog state tracking,” in Proceedings of … Cited by 2 Related articles All 2 versions

Optimizing human-interpretable dialog management policy using genetic algorithm H Ren, W Xu, Y Yan – 2015 IEEE Workshop on Automatic …, 2015 – ieeexplore.ieee.org … part in spoken dialog system (SDS) and its major functionalities include tracking dialog states … 6]. Firstly, RL algorithms are mostly data-demanding, which leaves dialog system designers in a … approach has been taken in evaluating the effect of different dialog state tracker on end … Cited by 1 Related articles All 4 versions

Multimodal addressee detection in multiparty dialogue systems TJ Tsai, A Stolcke, M Slaney – 2015 IEEE International …, 2015 – ieeexplore.ieee.org … 4 element linear microphone array and a wide-angle camera for visual tracking. … Lexical and dialog state information are also useful, providing significant performance gains … Shriberg, Andreas Stolcke, and Suman Ravuri, “Ad- dressee detection for dialog systems using temporal … Cited by 1 Related articles All 12 versions

Automatic detection of miscommunication in spoken dialogue systems R Meena, JLG Skantze… – 16th Annual Meeting of …, 2015 – anthology.aclweb.org … The corpus comprises of spoken interactions between the Cambridge Spoken Dialogue System and users, where the system provides restaurant rec- ommendations for Cambridge. The dialogue sys- tem is a research system that uses dialogue-state tracking for dialogue … Cited by 3 Related articles All 15 versions

Artificial conversations for customer service chatter bots: Architecture, algorithms, and evaluation metrics C Chakrabarti, GF Luger – Expert Systems with Applications, 2015 – Elsevier … State tracking is an important task in management of dialog systems. Several belief based state tracking architectures handle this problem using stochastic methods including generative and discriminative models (Deng et al., 2013). … Cited by 2 Related articles All 5 versions

Fusion paradigms in cognitive technical systems for human–computer interaction M Glodek, F Honold, T Geier, G Krell, F Nothdurft… – Neurocomputing, 2015 – Elsevier Recent trends in human–computer interaction (HCI) show a development towards cognitive technical systems (CTS) to provide natural and efficient operating prin. Cited by 8 Related articles All 5 versions

User behavior fusion in dialog management with multi-modal history cues M Yang, J Tao, L Chao, H Li, D Zhang, H Che… – Multimedia Tools and …, 2015 – Springer … 3 a One hundred three face tracking points; b The selected 29 facial points used … information but also their emotion and behavior, are needed synchronously consideration in dialog system. … 1) In BSDM, the user behavior and emotion changes contribute to dialog state transfer as … Cited by 1 Related articles All 6 versions

Pomdp based action planning and human error detection EMD Jean-Baptiste, P Rotshtein, M Russell – IFIP International Conference …, 2015 – Springer … 4.3 Belief State Update In the literature (eg, COACH system), the ADL is solved using the standard definitions of the MDP or POMDP. Here, a factored POMDP is implemented. The factorization of the POMDP is a technique used in dialogue systems [12, 13]. … Cited by 2 Related articles All 4 versions

Modern standards for VoiceXML in pervasive multimodal applications D Schnelle-Walka, S Radomski… – … Perspectives on the …, 2015 – books.google.com … User: The terrace. Most of these problems are shared by OwlSpeak (Heinroth, Denich, & Schmitt, 2010), which em- ploys an information state update (ISU) model (Larsson & Traum, 2000) reflecting the current dialog’s state as beliefs. … Cited by 1 Related articles All 3 versions

The MSIIP System for Dialog State Tracking Challenge 4 M Li, J Wu – colips.org … Finally, a slot-based score averaging method is used to build an ensemble tracker, which can help improving the performances of all single trackers. References … The dialog state tracking challenge. … Partially observable markov decision pro- cesses for spoken dialog systems. …

Dialogue State Tracking using Long Short Term Memory Neural Networks K Yoshino, T Hiraoka, G Neubig, S Nakamura – phontron.com … variety of expressions of users, because the users will be free from limitations im- posed by dialogue systems. … The dialogue state is affected by not only the features of the current utterance but also features of … We employed the LSTM implementation of the Pybrain1 in the tracker. … Related articles

Convolutional Neural Networks for Multi-topic Dialog State Tracking H Shi, T Ushio, M Endo, K Yamagami, N Horii – colips.org … Springer, 2012. 2. Seokhwan Kim, Luis Fernando D’Haro, Rafael E. Banchs, Jason Williams, and Matthew Hen- derson. The Fourth Dialog State Tracking Challenge. In Proceedings of the 7th International Workshop on Spoken Dialogue Systems (IWSDS), 2016. 3. Yoon Kim. …

A TV Program Discovery Dialog System Using Recommendations D Ramachandran, M Fanty, R Provine, PZ Yeh… – 16th Annual Meeting of …, 2015 – aclweb.org … In this version, we will additionally demonstrate the integration of the dialog system with a recom- mender engine that … For successive turns of the dialog, we use a be- lief tracking component that merges the REL-Tree for an input utterance with the dialog state, which is a … Related articles All 9 versions

Hybrid Dialog State Tracker M Vodolán, R Kadlec, J Kleindienst – arXiv preprint arXiv:1510.03710, 2015 – arxiv.org … It abstracts away the subsystems of end-to-end spoken dialog systems, focusing only on the … The last three dialog state tracking challenges [1, 2, 3] were dominated by machine learning based trackers [4 … We show that on the DSTC2 dataset our hybrid tracker achieves the state-of … Cited by 3 Related articles All 3 versions

Multi-agent learning in multi-domain spoken dialogue systems M Gašic, N Mrkšic, LR Barahona, PH Su, D Vandyke… – pdfs.semanticscholar.org … In future, we plan to apply this method in combination with a domain or a topic tracker operating over a large knowledge graph … [16] B Thomson and S Young, “Bayesian update of dialogue state: A POMDP framework for spoken dialogue systems,” Computer Speech and … Related articles All 3 versions

Analyzing dialogue breakdowns in chat-oriented dialogue systems R Higashinaka, K Funakoshi, M Mizukami… – errare2015.racai.ro … Many taxonomies of errors/miscommunication in dialogue have been created, espe- cially in task-oriented dialogue systems [13, 14, 15 … dataset to the public and hold an evaluation workshop (such as the Text REtrieval Conference [17] and Dialogue State Tracking Challenge [18 … Related articles

An Incremental Turn-Taking Model with Active System Barge-in for Spoken Dialog Systems T Zhao, AW Black, M Eskenazi – 16th Annual Meeting of the Special …, 2015 – aclweb.org … 2008. Towards incremental end-of-utterance detec- tion in dialogue systems. Proceedings of the 22nd International Conference on Computational Linguis- tics. … 2013. Dialog state tracking challenge. http://research. microsoft. com/en- us/events/dstc/. … Cited by 2 Related articles All 9 versions

A probabilistic framework for representing dialog systems and entropy-based dialog management through dynamic stochastic state evolution J Wu, M Li, CH Lee – IEEE/ACM Transactions on Audio, Speech …, 2015 – ieeexplore.ieee.org … 2 Spoken dialog system, probabilistic dialog representation, dialog state modeling, dialog management, automatic speech recognition, spoken language understanding, dialog turns, entropy minimization I. INTRODUCTION … Cited by 4 Related articles All 6 versions

Evaluation of Machine-Led Error Recovery Strategies for Domain Switches in a Spoken Dialog System S Reichel, U Ehrlich, A Berton, M Weber – … Language Dialog Systems and …, 2015 – Springer … limit the number of follow-up domains based on the current dialog state and partial … A, Rudnicky AI (2013) Predicting tasks in goal-oriented spoken dialog systems using semantic … OZ, Sarikaya R (2014) Hypotheses ranking for robust domain classification and tracking in dialogue … Related articles All 6 versions

Policy committee for adaptation in multi-domain spoken dialogue systems M Ga, N Mrk, PH Su, D Vandyke… – … IEEE Workshop on …, 2015 – ieeexplore.ieee.org … work we assume that the spoken language understand- ing unit, the belief tracker and the … [17] B Thomson and S Young, “Bayesian update of dialogue state: A POMDP … SJ Young, “Agenda-Based User Simulation for Boot- strapping a POMDP Dialogue System,” in Proceedings … Cited by 2 Related articles All 3 versions

YARBUS: Yet Another Rule Based belief Update System Jérémy Fix Hervé Frezza-Buet J Fix – 2015 – researchgate.net … The recently proposed dialog state tracking challenges offer the opportunity to test belief tracking algorithms on … The tracker proposed by Williams (2014) ranked first at the time of the … solve the belief tracking problem without requiring much of expert knowledge in dialog systems. … Related articles All 5 versions

Dialog Management with Deep Neural Networks L Zilka – pdfs.semanticscholar.org … The contribution of this thesis proposal is a novel approach to dialog state tracking. It aims towards building more responsive and simpler dialog systems by proposing the first trainable dialog state tracker which naturally operates incrementally, word-by-word, and can … Related articles All 3 versions

Belief Tracking with Stacked Relational Trees D Ramachandran, A Ratnaparkhi – 16th Annual Meeting of the Special …, 2015 – aclweb.org … The results show that the trained belief tracker outperformed the handcrafted on all measures, though … the first (to our knowledge) Belief Tracking approach that represents the dialog state with a … as making a marked difference in the task success rate of a complete dialog system. … Cited by 1 Related articles All 9 versions

Dialog Technology Bibliography NG Ward, G Skantze – pdfs.semanticscholar.org … Partially ob- servable markov decision processes for spoken dialog systems. Computer Speech & Language 21:393–422. Williams, JD; Henderson, M.; Raux, A.; Thomson, B.; Black, A.; and Ramachadran, D. 2014. The dialog state tracking challenge se- ries. … Related articles All 2 versions

[BOOK] Text, Speech, and Dialogue: 18th International Conference, TSD 2015, Pilsen, Czech Republic, September 14-17, 2015, Proceedings P Král, V Matoušek – 2015 – books.google.com … 131 Renáta Myšková and Petr Hájek Topic Classifier for Customer Service Dialog Systems….. … Catherine Middag, Florian Hönig, Jean-Pierre Martens, Michael Döllinger, Anne Schützenberger, and Elmar Nöth LecTrack: Incremental Dialog State Tracking with Long Short …

On the Granularity of Dialog Strategies: Insights from Large-scale Analyses of Two Commercial Travel Information Spoken Dialog Systems Z Jiao, Z Wang, G Wang, H Tian, H Wu, H Wang – 2015 – researchgate.net … Therefore, improving the granularity of dialog state representations and dialog strategy designs in travel … An empirical evaluation of a statistical dialog system in public use … Challenges and opportunities for state tracking in statistical spoken dialog systems: Results from two public … Related articles

Text, Speech, and Dialogue P Král, V Matoušek – Springer … 131 Renáta Myšková and Petr Hájek Topic Classifier for Customer Service Dialog Systems….. … Catherine Middag, Florian Hönig, Jean-Pierre Martens, Michael Döllinger, Anne Schützenberger, and Elmar Nöth LecTrack: Incremental Dialog State Tracking with Long Short …

Toward Multi-domain Language Generation using Recurrent Neural Networks TH Wen, M Gašic, N Mrkšic, LM Rojas-Barahona… – svr-ftp.eng.cam.ac.uk … However, the goal of building an open domain dialogue system that is capable of talking about any topic is still far away. Several works have begun to address this problem, such as [15] for multi-domain dialogue state tracking, or [16] for multi-domain dialogue management. … Related articles All 7 versions

Social talk capabilities for dialogue systems T Klüwer – 2015 – universaar.uni-saarland.de … Social Talk Capabilities for Dialogue Systems Tina Klüwer T in a K lü w e r S o cia l Ta lk C a p a b ilitie … Page 2. Tina Klüwer Social Talk Capabilities for Dialogue Systems universaar Universitätsverlag des Saarlandes Saarland University Press Presses Universitaires de la Sarre … Cited by 1 Related articles All 3 versions

Hyper-parameter Optimisation of Gaussian Process Reinforcement Learning for Statistical Dialogue Management L Chen, PH Su, M Gašic – 16th Annual Meeting of the Special …, 2015 – anthology.aclweb.org … 2015. Learning from real users: Rating di- alogue success with neural networks for reinforce- ment learning in spoken dialogue systems. Submit- ted to Interspeech. Kai Sun, Lu Chen, Su Zhu, and Kai Yu. 2014. A gener- alized rule based tracker for dialogue state tracking. … All 11 versions

Attention and Engagement Aware Multimodal Conversational Systems Z Yu – Proceedings of the 2015 ACM on International …, 2015 – dl.acm.org … [13] used gaze features obtained by an eye tracker to model … Thus we designed a coding scheme based on the knowledge of the open space the dialog system is positioned … We heuristically infer attention targets in real time based on the dialog state and the geo- metric direction … Cited by 1 Related articles All 2 versions

Easily Bootstrappable Statistical Spoken Dialogue System K Valev – 2015 – isl.anthropomatik.kit.edu … There are two broad approaches to achieving a practical and tractable implementation of a POMDP-based dialogue system. … The most obvious examples of these are the so-called slot filling applications where the complete dialogue state is reduced to the state of a small … Related articles

A proposal for improving spoken dialog systems using context information fusion I Chairi, D Griol, J García… – Information Fusion (Fusion …, 2015 – ieeexplore.ieee.org … for a harbor surveillance scenario, including rule- based reasoning to extend tracking data and … The spoken dialog system considers the concepts and values for the attributes provided by the … perceptron (MLP) [37], where the input layer holds the codification of the dialog state. … Related articles

Learning for Spoken Dialog Systems with Discriminative Graphical Models Y Ma – 2015 – etd.ohiolink.edu … 8 2.1 Overview of a Spoken Dialog System . . . . . 8 2.2 Previous Methods for Dialog State Tracking . . . . . 10 2.3 UnansweredChallengesforSpokenDialogSystems . . . . . 13 2.4 AnotherWaytoCombatOverfitting . . . . . 19 … Related articles All 3 versions

Measuring the differences between human-human and human-machine dialogs D Griol, J Molina – 2015 – gredos.usal.es … Our study has been performed using a dialog system called Facilisimo. … Large amounts of data are required for a systematic exploration of the dialog state space and corpora of simulated data are extremely valuable for this purpose. …

Utterance classification in speech-to-speech translation for zero-resource languages in the hospital administration domain LJ Martin, A Wilkinson, SS Miryala… – … IEEE Workshop on …, 2015 – ieeexplore.ieee.org … the search set. With data collected from a dialogue system we have de- veloped that uses only our predefined phrases, we plan on using dialogue-state tracking to improve performance of the recognition. For example, one would … Cited by 1 Related articles All 3 versions

A Study of Multimodal Addressee Detection in Human-Human-Computer Interaction TJ Tsai, A Stolcke, M Slaney – IEEE Transactions on Multimedia, 2015 – ieeexplore.ieee.org … [18] explore eye gaze, dialog state, and utterance … to keep in mind the limitations of the computer agent and the ongoing development of more natural dialogue systems. … Common eye trackers using a single camera without mechanical tracking might only be able to analyze the … Related articles All 6 versions

Context-Aware Automated Analysis and Annotation of Social Human–Agent Interactions T Baur, G Mehlmann, I Damian, F Lingenfelser… – ACM Transactions on …, 2015 – dl.acm.org … capabilities of the Microsoft Kinect SDK, as well as the Intraface Face Tracker [Xiong and … of the Eyetribe3 and SMI4 stationary eye trackers and Eye-Tracking Glasses (ETGs … support recording the agents’ behavior as well as meta information about the dialog state and discourse … Cited by 5 Related articles All 2 versions

Employing distance-based semantics to interpret spoken referring expressions I Zukerman, SN Kim, T Kleinbauer… – Computer Speech & …, 2015 – Elsevier In this paper, we present Scusi?, an anytime numerical mechanism for the interpretation of spoken referring expressions. Our contributions are: (1) an anytime i. Cited by 3 Related articles All 4 versions