RNN (Recurrent Neural Network) & Dialog Systems 2014

Recurrent Neural Network


See also:

100 Best Recurrent Neural Network VideosRNN (Recurrent Neural Network) & Dialog Systems 2015 | RNN (Recurrent Neural Network) & Question Answering Systems 2015

Word-based dialog state tracking with recurrent neural networks M Henderson, B Thomson… – 15th Annual Meeting of …, 2014 – anthology.aclweb.org … The method is based on a recurrent neural network structure which is capable of generalising to unseen dialog state hypotheses, and … 1 Introduction While communicating with a user, statistical spo- ken dialog systems must maintain a distribution over possible dialog states in a … Cited by 4 Related articles All 9 versions

ASR error detection using recurrent neural network language model and complementary ASR YC Tam, Y Lei, J Zheng, W Wang – Acoustics, Speech and …, 2014 – ieeexplore.ieee.org … may request rephrasing or spelling of words) [1]. In this way, accurate ASR error detection can play a crucial role within a dialogue system. We present two approaches to improve ASR error detection per- formance: (1) forward and backward recurrent neural network lan- guage … Cited by 3 Related articles All 5 versions

A historical perspective of speech recognition X Huang, J Baker, R Reddy – Communications of the ACM, 2014 – dl.acm.org … In Proceedings of Interspeech (2013). 38. Yao, K. et al. Recurrent neural networks for language understanding. … 42. Williams, J. and Young, S. Partially observable Markov decision processes for spoken dialog systems. Computer Speech and Language 21, 2 (2007), 393–422. … Cited by 9 Related articles

Feature enhancement by deep LSTM networks for ASR in reverberant multisource environments F Weninger, J Geiger, M Wöllmer, B Schuller… – Computer Speech & …, 2014 – Elsevier … Highlights. • Deep recurrent neural networks are used for data-based speech feature enhancement. • … 2. Feature enhancement. 2.1. Deep LSTM recurrent neural networks. In this article, we use deep LSTM recurrent neural networks (RNNs) for speech feature enhancement. … Cited by 7 Related articles All 4 versions

The Third Dialog State Tracking Challenge M Henderson, B Thomson… – Proceedings of IEEE …, 2014 – mi.eng.cam.ac.uk … [8] Matthew Henderson, Blaise Thomson, and Steve Young, “Word-Based Dialog State Tracking with Recurrent Neural Networks,” in Proceedings of … Steve Young, “Nat- ural actor and belief critic: Reinforcement algorithm for learning parameters of dialogue systems modelled as … Cited by 3 Related articles All 6 versions

Neural network language models for off-line handwriting recognition F Zamora-Martínez, V Frinken, S España-Boquera… – Pattern Recognition, 2014 – Elsevier … This work presents our latest contribution to this task, integrating neural network language models in the decoding process of three state-of-the-art systems: one based on bidirectional recurrent neural networks, another based on hybrid hidden Markov models and, finally, a … Cited by 4 Related articles All 4 versions

An artificial neural network approach to automatic speech processing SM Siniscalchi, T Svendsen, CH Lee – Neurocomputing, 2014 – Elsevier … Although several neural architectures have been proposed to generate state emission probabilities, such as recurrent neural networks (eg, [15] and [16]) and time-delay neural networks [17], the stylistic characteristics of the MLPs are by far the most popular due to the … Cited by 3 Related articles All 4 versions

Social signal classification using deep BLSTM recurrent neural networks R Brueckner, B Schulter – Acoustics, Speech and Signal …, 2014 – ieeexplore.ieee.org … and Dynamic Modelling for the Recognition of Non-Verbal Vocalisations in Conversational Speech,” in Perception in Multimodal Dialogue Systems: 4th IEEE … [16] A. Graves, A. rahman Mohamed, and G. Hinton, “Speech recognition with deep recurrent neural networks,” in Proc. … Cited by 1 Related articles All 5 versions

Inferring depression and affect from application dependent meta knowledge M Kächele, M Schels, F Schwenker – Proceedings of the 4th International …, 2014 – dl.acm.org … For the fusion of multi-modal signals in time continuous space, the use of recurrent neural networks has also become appealing. … kind of data collection is the EmoRec II corpus, where a subject is playing multiple rounds of a card game using a voice controlled dialog system [59 … Cited by 4 Related articles All 2 versions

Improved Factorization of a Connectionist Language Model for Single-Pass Real-Time Speech Recognition ? Brocki, D Koržinek, K Marasek – Foundations of Intelligent Systems, 2014 – Springer … In: Proceedings of the IIS 2008 Workshop on Spoken Language Understanding and Dialogue Systems (2008) 15. … Graves, A.: Sequence transduction with recurrent neural networks. In: CoRR. Vol- ume abs/1211.3711, Edinburgh, Scotland (2012) Cited by 1 Related articles All 3 versions

Reranked aligners for interactive transcript correction B Favre, M Rouvier, F Bechet – Acoustics, Speech and Signal …, 2014 – ieeexplore.ieee.org … tron trained on crowd-sourced examples, using various fea- tures such as probabilities from a Recurrent Neural Network Language Model … The task of improving automatic transcripts with interactions has mainly been pursued through confirmation in dialog systems and edition … Cited by 1 Related articles All 3 versions

A generalized rule based tracker for dialogue state tracking K Sun, L Chen, S Zhu, K Yu – submitted to IEEE SLT 2014, 2014 – aiexp.info … by the results of [1]. In contrast, dis- criminative state tracking models were successfully used for spoken dialogue systems [6]. The … includ- ing Maximum Entropy [12], Conditional Random Field [13], Deep Neural Network (DNN) [14], and Recurrent Neural Network [15] have been … Cited by 1 Related articles

Convolutional neural networks for speech recognition O Abdel-Hamid, AR Mohamed, H Jiang… – IEEE/ACM Transactions …, 2014 – dl.acm.org Page 1. IEEE/ACM TRANSACTIONS ON AUDIO, SPEECH, AND LANGUAGE PROCESSING, VOL. 22, NO. 10, OCTOBER 2014 1533 Convolutional Neural Networks for Speech Recognition Ossama Abdel-Hamid, Abdel-rahman … Cited by 5 Related articles All 7 versions

Inter-Annotator Agreement on Spontaneous Czech Language T Valenta, L Šmídl, J Švec, D Soutner – Text, Speech and Dialogue, 2014 – Springer … To record the corpus, a simple dialogue system was developed. … experiments we rescored the n-best list with recurrent neural network language model (RNN LM) [8] in combination with background LM which is a standard 5-gram smoothed with Kneser-Ney discounting [6]. The … Cited by 1 Related articles All 4 versions

Data collection and language understanding of food descriptions M Korpusik, N Schmidt, J Drexler, S Cyphers… – Proc. SLT, 2014 – groups.csail.mit.edu … [5] Y. Chen, W. Wang, and A. Rudnicky, “Unsupervised induction and filling of semantic slots for spoken dialogue systems using frame-semantic parsing,” in Proc. … 78–83. [7] G. Mesnil, X. He, L. Deng, and Y. Bengio, “Investigation of recurrent-neural-network architectures and … Cited by 1 Related articles All 4 versions

Application of deep belief networks for natural language understanding R Sarikaya, GE Hinton, A Deoras – IEEE/ACM Transactions on Audio, …, 2014 – dl.acm.org … all aspects of speech and lan- guage processing including natural language processing, spoken dialog systems, speech recognition … several decoding techniques for incorporating complex and long span language models, such as recurrent neural network language models, in … Cited by 3 Related articles All 7 versions

ASR Independent Hybrid Recurrent Neural Network Based Error Correction for Dialog System Applications J Choi, S Ryu, K Lee, Y Kim, S Koo, J Bang… – … enabling Artificial Agents …, 2014 – Springer Abstract We proposed an automatic speech recognition (ASR) error correction method using hybrid word sequence matching and recurrent neural network for dialog system applications. Basically, the ASR errors are corrected by the word sequence matching …

Robust Dialog State Tracking Using Delexicalised Recurrent Neural Networks And Unsupervised Adaptation M Henderson, B Thomson, S Young – mi.eng.cam.ac.uk … Before being de- ployed, most dialog systems are trained for well-defined and static domains such as this. … This paper presents how recurrent neural networks (RNNs) can be effectively applied to the task of dialog state tracking in expanding domains. … Related articles All 3 versions

Leveraging Frame Semantics and Distributional Semantics for Unsupervised Semantic Slot Induction in Spoken Dialogue Systems YN Chen, WY Wang, AI Rudnicky – cs.cmu.edu … The recurrent neural network language models use the context history to include long-distance information. … The purpose of the ranking model is to distinguish between generic semantic concepts and domain-specific concepts that are relevant to a spoken dialogue system. … Cited by 1 Related articles All 3 versions

Neural Network Models for Lexical Addressee Detection S Ravuri, A Stolcke – Fifteenth Annual Conference of the …, 2014 – mazsola.iit.uni-miskolc.hu … [1] E. Shriberg, A. Stolcke, and S. Ravuri, “Addressee de- tection for dialog systems using temporal and spectral di- mensions of speaking style”, in Proc. … 2000. [15] T. Mikolov, M. Karafiát, L. Burget, JH ?Cernocký, and S. Khudanpur, “Recurrent neural network based language … Related articles All 11 versions

Deep Neural Networks For Spoken Dialog Systems C MAIN – 2014 – macs.hw.ac.uk … Page 7. 1 Introduction 1.1 Motivation One of the major issues in Spoken Dialog Systems (SDS) is that the automatic speech recognition (ASR) and spoken language understanding (SLU) components can be error … 11 Page 12. network architecture known as the Hopfield Network. … Related articles

Deep Generative and Discriminative Models for Speech Recognition L Deng – wissap.iiit.ac.in Page 1. Deep Generative and Discriminative Models for Speech Recognition Li Deng deng@microsoft.com Microsoft Research, Redmond, WA 98052 WiSSAP-2014 Lecture (Janury 17-20) Page 2. Main References • Li Deng … Related articles

The Use Of Discriminative Belief Tracking In Pomdp-Based Dialogue Systems D Kim, M Henderson, M Gašic, P Tsiakoulis, S Young – svr-www.eng.cam.ac.uk … with the original Bayesian network tracker. Index Terms— dialogue management, spoken dialogue systems, recurrent neural networks, belief tracking, POMDP 1. INTRODUCTION Recent advances in statistical POMDP-based … Related articles All 3 versions

Joint Semantic Utterance Classification and Slot Filling with Recursive Neural Networks DZ Guo, G Tur, W Yih, G Zweig – research.microsoft.com … 3. RECURSIVE NEURAL NETWORKS FOR DIALOG SYSTEMS … Recurrent Neural Networks (RNNs) can be thought of as a limiting case of recursive neu- ral networks, where RNNs repeatedly apply a neural network to a degenerate tree (a chain) that has no notion of syntactic … Related articles All 6 versions

Interlocutor personality perception based on BFI profiles and coupled HMMs in a dyadic conversation MH Su, YT Zheng, CH Wu – Chinese Spoken Language …, 2014 – ieeexplore.ieee.org … First, the recurrent neural networks (RNNs) are adopted to project the linguistic features of the transcribed spoken text of the input speech … provide harmonious communication between humans and computers [3] [4] [5]. In order to endow the spoken dialogue systems with flexible … Related articles

Translation rescoring through recurrent neural network language models Á PERIS ABRIL – 2014 – riunet.upv.es … Recurrent neural network unfolded in time. … discourse analysis, machine translation, morphological segmentation, natural language generation and understanding, speech recognition, topic segmentation and recognition, information retrieval, dialog systems, question answering … Related articles

Structural information aware deep semi-supervised recurrent neural network for sentiment analysis W Rong, B Peng, Y Ouyang, C Li, Z Xiong – Frontiers of Computer Science – Springer … As a result a novel sentiment analysis model is pro- posed based on recurrent neural network, which takes the par- tial document as input and then the next parts to predict the sentiment label distribution rather than the next word. … 2.2 Recurrent neural network … Related articles

On a Hybrid NN/HMM Speech Recognition System with a RNN-Based Language Model D Soutner, J Zelinka, L Müller – Speech and Computer, 2014 – Springer … The recurrent neural networks were successfully introduced to the field of language modelling by Mikolov [9]. The basic scheme of … Present-day applications of the decoder include com- mercial dictation software, automatic subtitling system, dialogue system and systems for … Related articles All 3 versions

[BOOK] Statistical Language and Speech Processing L Besacier, AH Dediu, C Martín-Vide – 2014 – Springer … extraction; spelling correction; text and web mining; opinion mining and sentiment analysis; spoken dialog systems; author identification … Extraction and Categorization A Comparison of Sequence-Trained Deep Neural Networks and Recurrent Neural Networks Optical Modeling … All 4 versions

Cluster based Chinese Abbreviation Modeling Y Shi, YC Pan, MY Hwang – Fifteenth Annual Conference …, 2014 – mazsola.iit.uni-miskolc.hu … In some practical applications (eg dialogue systems and voice search systems), document level context information is not available. … Recurrent Neural Networks [14] demonstrates state-of-the- art performance in speech recognition [9, 15, 16], semantic analysis [17] and natural … Related articles All 4 versions

The Dialog State Tracking Challenge Series JD Williams, M Henderson, A Raux, B Thomson… – BE A PART OF …, 2014 – dropline.net … nine teams have participated in each DSTC, with global representation of the top research centers for spoken dialog systems. … the best-performing entries have been based on conditional random fields (Lee and Eskenazi 2013), recurrent neural networks (Henderson, Thomson … Related articles All 4 versions

Investigating Automatic & Human Filled Pause Insertion for Speech Synthesis R Dall, M Tomalin, M Wester… – … Annual Conference of …, 2014 – homepages.inf.ed.ac.uk … Finally we present results for initial at- tempts at FP insertion prediction using ngram LMs, a recurrent neural network (RRN) LM, support vector … pairs of sentences with and without FPs and were asked whether the FP increased the naturalness of a voice for a dialogue system. … Related articles All 7 versions

Neural network based feature extraction for speech and image recognition C Plahl – 2014 – darwin.bth.rwth-aachen.de … 107 7.6.2 Summary . . . . . 109 7.7 Stacking of Recurrent and Non-recurrent Neural Networks . . . . . … The recognized word sequence can be further processed by a machine translation system, a dialog system or any other text based system. … Related articles All 6 versions

The magazine archive includes every article published in Communications of the ACM for over the past 50 years. D Zhang – Communications of the ACM – cacm.acm.org … In Proceedings of Interspeech (2013). 38. Yao, K. et al. Recurrent neural networks for language understanding. … 42. Williams, J. and Young, S. Partially observable Markov decision processes for spoken dialog systems. Computer Speech and Language 21, 2 (2007), 393–422. … Related articles All 3 versions

Semantic Language Models For Automatic Speech Recognition AO Bayer, G Riccardi – sisl.disi.unitn.it … [8] T. Mikolov, M. Karafiat, L. Burget, J. Cernock, and S. Khudanpur, “Recurrent neural network based lan … Boquera, J. Castro- Bleda, M., and R. De-Mori, “Cache neural network language models based on long-distance dependencies for a spoken dialog system,” in Proceedings … Related articles All 2 versions

A Regression Approach to Speech Enhancement Based on Deep Neural Networks Y Xu, J Du, LR Dai, CH Lee – ieeexplore.ieee.org … Deep recurrent neural networks (DRNNs) were also adopted in the feature enhancement for robust speech recognition [24, 25]. The generalization capacity of the DRNN was weak if it was trained on limited noise types [24]. … Cited by 1 Related articles

Pattern Recognition for Biometrics and Bioinformatics KL Du, MNS Swamy – Neural Networks and Statistical Learning, 2014 – Springer … One of the key tasks of spoken-dialog systems is classification. Gait is an efficient biometric feature for human identification at a distance. … Inference of genetic regulatory networks with recurrent neural network models using particle swarm optimization. … Related articles

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 …, 2014 – Springer … effective multi-modal behavior fusion model and flexible behavior sensitive DM are necessary for practical human computer dialog systems. … It adopts recurrent neural network architectures which take into account past observations by cyclic connections in the network’s hidden … Cited by 1 Related articles All 2 versions

Intelligent Systems’ Holistic Evolving Analysis of Real-Life Universal Speaker Characteristics B Schuller, Y Zhang, F Eyben, F Weninger – mmk.e-technik.tu-muenchen.de … Com- bining static modelling of utterances with context knowl- edge, Long Short-Term Memory (LSTM) Recurrent Neural Networks (RNNs) have … spite them being crucial for real-life applications such as retrieval, dialogue systems and computer-mediated human- to-human … Related articles All 2 versions

Investigation of Speaker Group-Dependent Modelling for Recognition of Affective States from Speech I Siegert, D Philippou-Hübner, K Hartmann, R Böck… – Cognitive …, 2014 – Springer … Glüge S, Böck R, Wendemuth A. Segmented-memory recurrent neural networks versus hidden markov models in emotion recognition from speech. … Gnjatovi? M, Rösner D. On the role of the NIMITEK corpus in developing an emotion adaptive spoken dialogue system. … Related articles

Emotion Recognition and Depression Diagnosis by Acoustic and Visual Features: A Multimodal Approach M Sidorov, W Minker – Proceedings of the 4th International Workshop on …, 2014 – dl.acm.org … Such opportunity can be useful in various applications, eg, improvement of Spoken Dialogue Systems (SDSs) or mon- itoring agents in call-centers. … Real-life voice activity detection with lstm recurrent neural networks and an application to hollywood movies. … Cited by 1 Related articles

Automatic speech recognition for under-resourced languages: A survey L Besacier, E Barnard, A Karpov, T Schultz – Speech Communication, 2014 – Elsevier Speech processing for under-resourced languages is an active field of research, which has experienced significant progress during the past decade. We propose, i. Cited by 21 Related articles All 6 versions

State of Research of Speech Recognition M Sarma, KK Sarma – … -Based Speech Segmentation using Hybrid Soft …, 2014 – Springer … [ 69 ] where a spoken dialog system is designed to use in agricultural commodities task … Doctoral thesis, University of Cambridge, Cambridge. 38. Ahmad AM, Ismail S, Samaonl DF (2004) Recurrent neural network with backpropagation through time for speech recognition. … Related articles

Horn And Whistle Recognition Tech-niques For NAO Robots NW Backer, A Visser – Bachelor thesis, Universiteit van Amsterdam, 2014 – staff.fnwi.uva.nl Page 1. Horn And Whistle Recognition Tech- niques For NAO Robots Niels W. Backer 6082025 niels.backer@student.uva.nl Bachelor thesis Credits: 6 EC Bachelor Opleiding Kunstmatige Intelligentie University of Amsterdam … Cited by 2 Related articles All 2 versions

Language Learning via Unsupervised Corpus Analysis B Goertzel, C Pennachin, N Geisweiller – Engineering General Intelligence …, 2014 – Springer … and affect the recognition of lower level patterns. Our approach does not use conventional deep learning archi- tectures like Deep Boltzmann machines or recurrent neural networks. Conceptually, our approach is based on a …

Attentional Mechanisms for Socially Interactive Robots–A Survey JF Ferreira, J Dias – Autonomous Mental Development, IEEE …, 2014 – ieeexplore.ieee.org … [63]. Following a similar line of interest, Ikegami and Iizuka [90] proposed a coupled dynamical recogniser as a model for simulating turn-taking behaviour. A recurrent neural network Page 7. 1943-0604 (c) 2013 IEEE. Personal … Cited by 2 Related articles All 2 versions

Computer Aided Pronunciation Learning For Al-Jabari Method: A Review NJ Ibrahim, MYI Idris, Z Mohd Yusoff – 2014 – works.bepress.com … referring to the new achievements in the fields of computer aided learning systems, intelligent tutoring systems, speech recognition, speech synthesis and dialogue systems, which permits … Recurrent Neural Network with Backpropagation through Time for Speech Recognition. … Related articles All 3 versions

Foundations and Trends in Signal Processing L Deng, Y Dong – Signal Processing, 2014 – research.microsoft.com … the recognizer. More recently, a recurrent neural network (RNN) is used as the “backend” recognizer receiving the high-dimensional, DNN-derived features as the input without dimensionality reduction [48, 85]. These studies … Related articles All 8 versions

Cross-Domain and Cross-Language Porting of Shallow Parsing EA Stepanov – 2014 – eprints-phd.biblio.unitn.it … We have contributed to the optimization with the re-ranking of n-best list of translation hypotheses with the joined Recurrent Neural Network Language Model trained on reference word-concept pairs. An alternative to the cross-language porting of language resource anno- … Related articles

Modeling language with structured penalties AK Nelakanti – 2014 – tel.archives-ouvertes.fr … interactive machines. It has applications spanning across various domains, such as dialogue systems, text generation and machine translation among others and has been studied extensively in the past decades. Among problems … Related articles All 6 versions

Developmental reasoning and planning with robot through enactive interaction with human M Petit – 2014 – tel.archives-ouvertes.fr … link between a sentence and its meaning. This method has been implemented with a recurrent neural network, using a database provided from the human by interaction with the robot. The control of the language allows the … Related articles All 3 versions

Latent semantic rational kernels for topic spotting on conversational speech C Weng, DL Thomson, P Haffner… – IEEE/ACM Transactions on …, 2014 – dl.acm.org Page 1. 1738 IEEE/ACM TRANSACTIONS ON AUDIO, SPEECH, AND LANGUAGE PROCESSING, VOL. 22, NO. 12, DECEMBER 2014 Latent Semantic Rational Kernels for Topic Spotting on Conversational Speech Chao Weng … Related articles All 2 versions