LSTM (Long Short Term Memory) & Dialog Systems 2015


Notes:

Long short-term memory (LSTM) is a recurrent neural network (RNN) architecture, an artificial neural network (ANN).

Resources:

Wikipedia:

See also:

100 Best Recurrent Neural Network VideosAutonomous Agents, Natural Language & Dialog Systems 2015Classifier & Dialog SystemsCNN (Convolutional Neural Network) & Natural Language 2014Conditional Random Fields & Dialog Systems 2013Deep Belief Network & Dialog SystemsNeural Language ModelsNeural Network & Dialog SystemsNeural Network & Dialog Systems 2014Paraphrase DatabaseRNN (Recurrent Neural Network) & Dialog Systems 2014Streaming Data & Dialog SystemsVisual Question Answering


Semantically conditioned lstm-based natural language generation for spoken dialogue systems TH Wen, M Gasic, N Mrksic, PH Su, D Vandyke… – arXiv preprint arXiv: …, 2015 – arxiv.org … This pa- per presents a statistical language gener- ator based on a semantically controlled Long Short-term Memory (LSTM) struc- ture … The natural language generation (NLG) compo- nent provides much of the persona of a spoken dialogue system (SDS), and it has a significant … Cited by 51 Related articles All 18 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. … Page 2. ment frequency (TF-IDF) approach, to more so- phisticated neural models including a Recurrent Neural Network (RNN) and a Long Short-Term Memory (LSTM) architecture. … Cited by 20 Related articles All 12 versions

Neural responding machine for short-text conversation L Shang, Z Lu, H Li – arXiv preprint arXiv:1503.02364, 2015 – arxiv.org … 1997. Long short-term memory. Neural computation, 9(8):1735–1780. [Howell2010] David C. Howell. 2010. … 2000. Njfun: a reinforce- ment learning spoken dialogue system. In Proceedings of the 2000 ANLP/NAACL Workshop on Conversational systems, pages 17–20. ACL. … Cited by 54 Related articles All 12 versions

A hierarchical neural autoencoder for paragraphs and documents J Li, MT Luong, D Jurafsky – arXiv preprint arXiv:1506.01057, 2015 – arxiv.org … In this paper, we explore an important step toward this generation task: training an LSTM (Long- short term memory) auto-encoder to pre- serve and reconstruct multi-sentence para- graphs. … 2 Long-Short Term Memory (LSTM) … Cited by 40 Related articles All 15 versions

A diversity-promoting objective function for neural conversation models J Li, M Galley, C Brockett, J Gao, B Dolan – arXiv preprint arXiv: …, 2015 – arxiv.org … Earlier efforts to incorporate statistical methods into dialog systems typically relied on one of two ap- proaches. … Page 3. are Long Short-Term Memory (LSTM) neural net- works (Hochreiter and Schmidhuber, 1997) that are able to implicitly capture compositionality and long- span … Cited by 25 Related articles All 5 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 … Short-Term Memory (LSTMs) (Hochreiter & Schmidhuber, 1997), and attention-based models, in particular Memory Networks (Sukhbaatar et al., 2015). 2 THE MOVIE DIALOG DATASET We introduce a set of four tasks to test the ability of end-to-end dialog systems, focusing on … Cited by 10 Related articles All 3 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

Neural generative question answering J Yin, X Jiang, Z Lu, L Shang, H Li, X Li – arXiv preprint arXiv:1512.01337, 2015 – arxiv.org … generating relevant and fluent responses in natural language in a conver- sation [5, 6]. Recent progress in neural dialogue system has raised … extended the current scheme of representing text information in both short-term memory (eg, in [1]) and long short-term memory (eg, in … Cited by 6 Related articles All 4 versions

Improved deep learning baselines for ubuntu corpus dialogs R Kadlec, M Schmid, J Kleindienst – arXiv preprint arXiv:1510.03753, 2015 – arxiv.org … [1] R. Lowe, N. Pow, I. Serban, and J. Pineau, “The ubuntu dialogue corpus: A large dataset for research in unstructured multi-turn dialogue systems,” arXiv preprint arXiv:1506.08909, 2015. … [4] S. Hochreiter and J. Schmidhuber, “Long Short-Term Memory,” Neural Computation … Cited by 6 Related articles All 3 versions

A critical review of recurrent neural networks for sequence learning ZC Lipton, J Berkowitz, C Elkan – arXiv preprint arXiv:1506.00019, 2015 – arxiv.org … In recent years, systems based on long short-term memory (LSTM) and bidirectional (BRNN) architectures have demonstrated ground-breaking … Besides dialogue systems, modern interactive systems of economic importance include self-driving cars and robotic surgery, among … Cited by 24 Related articles All 10 versions

Speech based emotion recognition V Sethu, J Epps, E Ambikairajah – Speech and Audio Processing for …, 2015 – Springer … M. Wollmer, B. Schuller, F. Eyben, G. Rigoll, Combining long short-term memory and dynamic Bayesian networks for incremental emotion-sensitive artificial … R. Tato, T. Kemp, B. Meffert, Towards real life applications in emotion recognition, in Affective Dialogue Systems, ed. by E … Cited by 8 Related articles All 4 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 … 2015. Reinforcement-learning based dialogue system for human-robot interactions with socially-inspired rewards. Computer Speech & Language, 34(1):256–274. … 1997. Long short-term memory. Neural computation, 9(8):1735–1780. … Cited by 2 Related articles All 18 versions

LecTrack: Incremental Dialog State Tracking with Long Short-Term Memory Networks L Žilka, F Jurcícek – International Conference on Text, Speech, and …, 2015 – Springer … dialog state tracker is an important component in modern spoken dialog systems. We present the first trainable incremental dialog state tracker that directly uses automatic speech recognition hypotheses to track the state. It is based on a long short-term memory recurrent neural … Cited by 1 Related articles

Is it time to switch to Word Embedding and Recurrent Neural Networks for Spoken Language Understanding? V Vukotic, C Raymond, G Gravier – InterSpeech, 2015 – hal.inria.fr … in [4, 5] where they show, on the popular ATIS database, that Recurrent Neu- ral Networks and Long Short Term Memory Neural Networks … 2.2. MEDIA The research project MEDIA [8] evaluates different SLU mod- els of spoken dialogue systems dedicated to provide tourist in … Cited by 12 Related articles All 11 versions

Learning to understand phrases by embedding the dictionary F Hill, K Cho, A Korhonen, Y Bengio – arXiv preprint arXiv:1504.00548, 2015 – arxiv.org … At = f(UAt-1 + Wvt + b). As such, the values of the final internal activation state units AN are a weighted function of all input word embeddings, and constitute a ‘summary’ of the information in the sentence. 2.1 Long Short Term Memory … Cited by 12 Related articles All 10 versions

Advances in natural language processing J Hirschberg, CD Manning – Science, 2015 – science.sciencemag.org … Today’s researchers refine and make use of such tools in real-world applications, creating spoken dialogue systems and speech … translation, research has focused on a particular version of recurrent neural networks, with enhanced “long short-term memory” computational units … Cited by 22 Related articles All 9 versions

Incremental LSTM-based dialog state tracker L Zilka, F Jurcicek – 2015 IEEE Workshop on Automatic Speech …, 2015 – ieeexplore.ieee.org … [7] Ondrej Dušek, Ondrej Plátek, Lukáš Žilka, and Filip Ju- rcícek, “Alex: Bootstrapping a spoken dialogue system for a … 22, no. 5, pp. 16–31, 2005. [10] Sepp Hochreiter and Jürgen Schmidhuber, “Long short- term memory,” Neural computation, vol. 9, no. 8, pp. 1735–1780, 1997. … Cited by 6 Related articles All 4 versions

Recurrent neural network and LSTM models for lexical utterance classification S Ravuri, A Stolcke – Proc. Interspeech, Dresden, 2015 – research-srv.microsoft.com … [2] E. Shriberg, A. Stolcke, and S. Ravuri, “Addressee de- tection for dialog systems using temporal and spectral di- mensions of speaking style”, in Proc. Interspeech, Lyon, Aug. 2013. … [10] S. Hochreiter and J. Schmidhuber, “Long short-term memory”, Neural Comput., vol. 9, pp. … Cited by 8 Related articles All 5 versions

Listen, attend, and walk: Neural mapping of navigational instructions to action sequences H Mei, M Bansal, MR Walter – arXiv preprint arXiv:1506.04089, 2015 – arxiv.org … Our alignment-based encoder-decoder model with long short-term memory recurrent neural net- works (LSTM-RNN) translates natural language instructions to action sequences based upon a representation of the ob- servable world state. … Cited by 8 Related articles All 8 versions

Efficient learning for spoken language understanding tasks with word embedding based pre-training Y Luan, S Watanabe, B Harsham – Sixteenth Annual Conference of the …, 2015 – Citeseer … Re- search and commercial spoken dialog systems,” in Proceedings of the 6th Annual Meeting of the Special Interest Group on Discourse and … Yao, B. Peng, Y. Zhang, D. Yu, G. Zweig, and Y. Shi, “Spo- ken language understanding using long short-term memory neural networks … Cited by 4 Related articles All 8 versions

Contextual spoken language understanding using recurrent neural networks Y Shi, K Yao, H Chen, YC Pan… – … on Acoustics, Speech …, 2015 – ieeexplore.ieee.org … consists of domain identification, intent classification and slot filling [1]. SLU is a crit- ical component in spoken dialogue systems. … using the proposed RNN-based joint training method, in comparison to the published results using CRF, RNN, Long Short-Term-Memory (LSTM) [28 … Cited by 11 Related articles All 7 versions

Relevance units machine based dimensional and continuous speech emotion prediction F Wang, H Sahli, J Gao, D Jiang, W Verhelst – Multimedia Tools and …, 2015 – Springer … [38, 39], and Nicolaou et al. [22], investigated techniques based on context modeling using Long Short-Term Memory neural networks (LSTM). These systems provide the advantage to encode the long-range dependencies between observations within the learning algorithm. … Cited by 5 Related articles All 3 versions

Discriminative methods for statistical spoken dialogue systems MS Henderson – 2015 – repository.cam.ac.uk Page 1. Discriminative Methods for Statistical Spoken Dialogue Systems Matthew S. Henderson … It has been a privilege to work with my lab-mates in the Cambridge Dialogue Systems group, whose help in developing the work presented here has been invaluable. I would … Cited by 4 Related articles All 3 versions

Attention with Intention for a Neural Network Conversation Model K Yao, G Zweig, B Peng – arXiv preprint arXiv:1510.08565, 2015 – arxiv.org … 3.5 Implementation details All of the recurrent networks are implemented using a recently proposed depth-gated long-short- term memory (LSTM) network [16]. … [18] S. Young, M. Gasic, B. Thomson, and JD Williams. POMDP-based statistical spoken dialog systems: A review. … Cited by 8 Related articles All 3 versions

Machine learning for dialog state tracking: A review M Henderson – 2015 – research.google.com … Early spoken dialog systems used hand-crafted rules for DST, keep- ing a single top hypothesis for each component of the dialog state … Long Short-Term Memory (LSTM) networks [42] have also been applied to the task of word-based DST, modeling not only the sequence of … Cited by 3 Related articles All 2 versions

Modeling phrasing and prominence using deep recurrent learning A Rosenberg, R Fernandez… – … Annual Conference of …, 2015 – researchgate.net … [12] R. Fernandez, A. Rendel, B. Ramabhadran, and R. Hoory, “Prosody contour prediction with long short-term memory, bi- directional … Heldner, “An instantaneous vec- tor representation of delta pitch for speaker-change prediction in conversational dialogue systems,” in ICASSP … Cited by 6 Related articles All 2 versions

Recurrent Reinforcement Learning: A Hybrid Approach X Li, L Li, J Gao, X He, J Chen, L Deng, J He – arXiv preprint arXiv: …, 2015 – arxiv.org … First, unlike Mnih et al. (2015), we employ recurrent neural networks (RNN) and long short-term memory (LSTM) (Hochreiter and Schmidhuber 1997) models to learn the representation of states for RL. … Reinforcement learning with long short-term memory. … Cited by 3 Related articles All 3 versions

Towards universal paraphrastic sentence embeddings J Wieting, M Bansal, K Gimpel, K Livescu – arXiv preprint arXiv: …, 2015 – arxiv.org … We present the interesting result that simple com- positional architectures based on updated vector averaging vastly outperform long short-term memory (LSTM) recurrent neural networks and that these simpler ar- chitectures allow us to learn models with superior generalization … Cited by 13 Related articles All 2 versions

A comparative study of neural network models for lexical intent classification S Ravuri, A Stoicke – 2015 IEEE Workshop on Automatic …, 2015 – ieeexplore.ieee.org … models. 1. INTRODUCTION Utterance classification is an important pre-processing step for many dialog systems that interpret speech input. For … network and long short-term memory units [2] for both addressee and intent detection. Most … Cited by 1 Related articles All 4 versions

Deep Contextual Language Understanding in Spoken Dialogue Systems C Liu, P Xu, R Sarikaya – Sixteenth Annual Conference of …, 2015 – research.microsoft.com … Moreover, adding dialog system response as external features provides consistent further gains, being complementary with recurrent SLU features. … B. Peng, Y. Zhang, D. Yu, G. Zweig, and Y. Shi, “Spo- ken language understanding using long short-term memory neural networks … Cited by 2 Related articles All 4 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 … An effective multi-modal behavior fusion model and flexible behavior sensitive DM are necessary for practical human computer dialog systems. … In term of the multi-modal behavior recognition, Long Short-Term memory (LSTM) method has been proved to be a successful model … Cited by 1 Related articles All 6 versions

Human Affect Recognition: Audio-Based Methods B Schuller, F Weninger – Wiley Encyclopedia of Electrical and …, 2015 – Wiley Online Library … is hard, particularly the learning of long-range dependencies (35), more advanced RNN architectures such as long short-term memory (36) have … disorders, from healthy adults to adults with voice pathologies, from interactions with spoken language dialog systems to emotional … Cited by 1 Related articles

Exploring the body and head kinematics of laughter, filled pauses and breaths S Kousidis, J Hough, D Schlangen – Proceedings of the 4th …, 2015 – dsg-bielefeld.de … Spyros Kousidis, Julian Hough and David Schlangen Dialogue Systems Group, Bielefeld University, Germany spyros.kousidis@uni-bielefeld.de ABSTRACT … Audiovisual Classification Of Vocal Outbursts In Human Conversation Using Long-Short-Term-Memory Networks. … Cited by 1 Related articles All 5 versions

Deep Reinforcement Learning with an Unbounded Action Space J He, J Chen, X He, J Gao, L Li, L Deng… – arXiv preprint arXiv: …, 2015 – arxiv.org … Narasimhan et al. (2015) applied an LSTM (long short term memory) -DQN framework to the task of learning control policies for parser-based text games, which achieves higher average reward than the random and BOW (bag-of-words) -DQN baselines. … Cited by 4 Related articles All 3 versions

Dialogue State Tracking using Long Short Term Memory Neural Networks K Yoshino, T Hiraoka, G Neubig, S Nakamura – phontron.com … 3. Sepp Hochreiter and Jürgen Schmidhuber. Long short-term memory. Neural computation, 9(8):1735–1780, 1997. … The Fourth Dialog State Tracking Challenge. In Proceedings of the 7th International Workshop on Spoken Dialogue Systems (IWSDS), 2016. … Related articles

Context Sensitive Spoken Language Understanding using Role Dependent LSTM layers H Chiori, T Hori, S Watanabe, JR Hershey – 2015 – pdfs.semanticscholar.org … short-term memory neural networks,” in Spoken Language Technology Workshop (SLT), December 2014. [6] Chiori Hori, Kiyonori Ohtake, Teruhisa Misu, Hideki Kashioka, and Satoshi Nakamura, “Sta- tistical dialog management applied to WFST-based dialog systems,” in IEEE … Cited by 1 Related articles All 3 versions

Dialog Management with Deep Neural Networks L Zilka – pdfs.semanticscholar.org … In this section we will describe our dialog state tracking model based on the long short-term memory units. Our model fits in the big picture of a spoken dialog system described in subsection 2.1 as a spoken language understanding and dialog state tracker together. … Related articles All 3 versions

Toward Multi-domain Language Generation using Recurrent Neural Networks TH Wen, M Gašic, N Mrkšic, LM Rojas-Barahona… – svr-ftp.eng.cam.ac.uk … We found that by imposing a sigmoid gate on the dialogue act vector, the Semantically Conditioned Long Short-term Memory generator can … significant progress has been made in applying statistical methods to auto- mate the development of Spoken Dialogue Systems (SDS) [ … Related articles All 7 versions

Natural Language Dialogue-Future Way of Accessing Information H Li – 2015 – hangli-hl.com … 2015) Page 40. Natural Language Dialogue System – Retrieval based Approach index of messages and responses matching ranking message … handsome boy Page 45. Natural Language Dialogue System – Generation based Approach • Encoding messages to intermediate … Related articles All 2 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 dialog state tracking. … 5 Page 6. [13] S. Hochreiter and J. Schmidhuber, “Long Short-Term Memory,” Neural Computation, vol. 9, pp. 1735–1780, 1997. … Cited by 3 Related articles All 3 versions

Emotion recognition in spontaneous and acted dialogues L Tian, JD Moore, C Lai – Affective Computing and Intelligent …, 2015 – ieeexplore.ieee.org … Both Support Vector Machines (SVM) and Long Short-Term Memory Recurrent Neural Net- works (LSTM-RNN) were built using each feature set on … of emotion is crucial for advancing technologies related to human-computer interaction, such as human-agent dialogue systems. … Cited by 1 Related articles All 10 versions

An Auto-Encoder for Learning Conversation Representation Using LSTM X Zhou, B Hu, Q Chen, X Wang – International Conference on Neural …, 2015 – Springer … First, the long short term memory (LSTM) neural network is used to encode the sequence of sentences in a conversation. … Linguist. 26(3), 339–374 (2000)CrossRef. 4. Rieser, V., Lemon, O.: Natural language generation as planning under uncertainty for spoken dialogue systems. … Related articles

Recognizing emotions in dialogues with acoustic and lexical features L Tian, JD Moore, C Lai – Affective Computing and Intelligent …, 2015 – ieeexplore.ieee.org … 20] ? Statistical: Low-Level Descriptors (LLD) [21] • Model: ? Contextual: Non-contextual:Support Vector Machine (SVM) Contextual: Long Short-Term Memory Recurrent Neural … We plan to work on this question in the future if we have an available dialogue system to apply our … Related articles All 6 versions

Recurrent Models for Auditory Attention in Multi-Microphone Distance Speech Recognition S Kim, I Lane – arXiv preprint arXiv:1511.06407, 2015 – arxiv.org … Many real-world speech recognition applications, including teleconferencing, robotics and in-car spoken dialog systems, must deal with speech from … This ^X is used for the next subnetwork LSTM-AM, which is a Long Short-Term Memory (LSTM) acoustic model to estimate the … Cited by 1 Related articles All 3 versions

Semi-supervised slot tagging in spoken language understanding using recurrent transductive support vector machines Y Shi, K Yao, H Chen, YC Pan… – 2015 IEEE Workshop on …, 2015 – ieeexplore.ieee.org … RNN. To overcome this issue, advanced RNN based slot tagging methods, for ex- ample long short term memory (LSTM) networks [21, 22], gated RNN [11] and RNN with external memory (RNNEM) [11] have been proposed. … Related articles

A language model based approach towards large scale and lightweight language identification systems BML Srivastava, HK Vydana, AK Vuppala… – arXiv preprint arXiv: …, 2015 – arxiv.org … module for a wide range of mul- tilingual applications like, call centers, multilingual Spoken Dialog Systems, emergency services … Various approaches featuring Feed Forward Deep Neural Networks (FF-DNNs) and Long-Short Term Memory Recurrent Neural Networks (LSTM … Related articles All 3 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….. … Hönig, Jean-Pierre Martens, Michael Döllinger, Anne Schützenberger, and Elmar Nöth LecTrack: Incremental Dialog State Tracking with Long Short-Term Memory Networks …

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

Multi-domain dialogue success classifiers for policy training D Vandyke, PH Su, M Gasic, N Mrksic… – … IEEE Workshop on …, 2015 – ieeexplore.ieee.org … Tsung-Hsien Wen, and Steve Young, “Learning from real users: Rating dialogue suc- cess with neural networks for reinforcement learning in spoken dialogue systems,” in Interspeech … 10] Sepp Hochreiter and Jürgen Schmidhuber, “Long short- term memory,” Neural computation … Cited by 3 Related articles All 8 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 … Rao, Fuchun Peng, Hasim Sak, and Françoise Bea- ufays, “Grapheme-to-phoneme conversion using long short- term memory recurrent neural … Franco, “Using dialog corrections to improve speech recognition,” in Error Handling in Spoken Language Dialogue Systems, 2003. … Cited by 5 Related articles All 6 versions

Towards Universal Paraphrastic Sentence Embeddings JWMBK Gimpel, K Livescu – arXiv preprint arXiv: …, 2015 – pdfs.semanticscholar.org … We present the interesting result that simple com- positional architectures based on updated vector averaging vastly outperform long short-term memory (LSTM) recurrent neural networks and that these simpler ar- chitectures allow us to learn models with superior generalization … Related articles All 4 versions

Video paragraph captioning using hierarchical recurrent neural networks H Yu, J Wang, Z Huang, Y Yang, W Xu – arXiv preprint arXiv:1510.07712, 2015 – arxiv.org … state for the sentence generator. The RNNs exploited by the two generators incorporate the Gated Recurrent Unit (GRU) [9] which is a simplification of the Long Short-Term Memory (LSTM) architecture [16]. In the following, we … Cited by 8 Related articles All 5 versions

Deep Reinforcement Learning with an Action Space Defined by Natural Language J He, J Chen, X He, J Gao, L Li… – arXiv preprint arXiv …, 2015 – pdfs.semanticscholar.org … Narasimhan et al. (2015) applied a Long Short-Term Memory DQN framework to the task of learning control policies for parser-based text games, which achieves higher average reward than the random and Bag-of-Words DQN baselines. … Cited by 4 Related articles

The SENSEI Project: Making Sense of Human Conversations G Riccardi, F Bechet, M Danieli, B Favre… – … Workshop on Future …, 2015 – Springer … the joint optimization of DNN over several NLP tasks; the ability of Recurrent Neural Networks (RNN) to maintain contextual information through sequence decoding with a memory model such as the Long Short Term Memory model [55]. … Related articles All 2 versions

Using recurrent neural networks for slot filling in spoken language understanding G Mesnil, Y Dauphin, K Yao, Y Bengio… – … on Audio, Speech, …, 2015 – ieeexplore.ieee.org Page 1. 2329-9290 (c) 2013 IEEE. Personal use is permitted, but republication/ redistribution requires IEEE permission. See http://www.ieee.org/ publications_standards/publications/rights/index.html for more information. This … Cited by 60 Related articles All 16 versions

Semi-Autonomous Data Enrichment and Optimisation for Intelligent Speech Analysis Z Zhang – 2015 – mediatum.ub.tum.de … For ASR, promising context-sensitive networks, namely Long Short- Term Memory (LSTM) neural networks [17], are employed to advance the feature enhancement … Generally, a Wizard-of-Oz (WOZ) is designed at the beginning (eg, a multi-model dialogue system), then elicits … Cited by 1 Related articles All 3 versions

Culture Clubs Processing Speech by Deriving and Exploiting Linguistic Subcultures DG Brizan – 2015 – gc.cuny.edu … as a tool for actors, it is become valuable to customer service representatives and people in other fields. A dialogue system could be created from the speech signals of one linguistic subculture and could generate prompts to the human interlocutor using text-to-speech (TTS). … Related articles All 2 versions

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, 2015 – Springer Page 1. Front. Comput. Sci., 2015, 9(2): 171–184 DOI 10.1007/s11704-014-4085-7 Structural information aware deep semi-supervised recurrent neural network for sentiment analysis Wenge RONG1,2, Baolin PENG1, Yuanxin OUYANG 1,2, Chao LI1,2, Zhang XIONG1,2 … Cited by 2 Related articles All 4 versions

A system for recognizing human emotions based on speech analysis and facial feature extraction: applications to Human-Robot Interaction M Rabiei – 2015 – dspace-uniud.cineca.it … emotion recognition has also been used in call center applications and mobile communication [2]. Some works tried to incorporate spoken dialogue system technology and service robots. Psychologists … detecting linguistic keyword features and long/short-term memory (LSTM). … Related articles All 2 versions

Sentiment analysis: Detecting valence, emotions, and other affectual states from text SM Mohammad – Emotion Measurement, 2015 – books.google.com … Brand management, customer relationship management, and stock market: Sentiment analysis of blogs, tweets, and Facebook posts is already widely used to shape brand image, track customer response, and in developing automatic dialogue systems for handling cus … Cited by 15 Related articles All 4 versions

Recurrent Neural Networks in Speech Disfluency Detection and Punctuation Prediction M Reisser – 2015 – isl.anthropomatik.kit.edu … in- creasingly relevant. These applications, such as automated machine transla- tion systems, dialogue systems or information extraction systems, usually are trained on large amount of text corpora. Since acquiring, manually …

MERL Annual Report 2015 RC Waters – 2015 – pdfs.semanticscholar.org Page 1. MITSUBISHI ELECTRIC RESEARCH LABORATORIES http://www.merl.com MERL Annual Report 2015 Waters, RC TR2015-000 June 2015 Abstract Welcome to Mitsubishi Electric Research Laboratories (MERL), the …

Nastalique segmentation-based approach for Urdu OCR S Hussain, S Ali – International Journal on Document Analysis and …, 2015 – Springer … These features along with character labeling are used by classifier for the training and recognition. Sankaran and Jawahar [ 15 ] use bidirectional Long-Short Term Memory (LSTM) model of Neural Networks for recognition of Devanagari text. … Related articles All 3 versions

Data-driven deep-syntactic dependency parsing M BALLESTEROS, B BOHNET… – Natural Language …, 2015 – Cambridge Univ Press Page 1. Natural Language Engineering: page 1 of 36. c Cambridge University Press 2015 doi:10.1017/S1351324915000285 1 Data-driven deep-syntactic dependency parsing† MIGUEL BALLESTEROS1, BERND BOHNET2 … Cited by 2 Related articles All 3 versions