Text, Speech, and Dialogue: 16th International Conference, TSD 2013, Pilsen, Czech Republic, September 1-5, 2013, Proceedings (Lecture Notes in Computer Science, 8082)

This book constitutes the refereed proceedings of the 16th International Conference on Text, Speech and Dialogue, TSD 2013, held in Pilsen, Czech Republic, in September 2013. The 65 papers presented together with 5 invited talks were carefully reviewed and selected from 148 submissions. The main topics of this year's conference was corpora, texts and transcription, speech analysis, recognition and synthesis, and their intertwining within NL dialogue systems. The topics also included speech recognition, corpora and language resources, speech and spoken language generation, tagging, classification and parsing of text and speech, semantic processing of text and speech, integrating applications of text and speech processing, as well as automatic dialogue systems, and multimodal techniques and modelling. Front Matter....Pages - Corpora of the Russian Language....Pages 1-13 Long, Deep and Wide Artificial Neural Nets for Dealing with Unexpected Noise in Machine Recognition of Speech....Pages 14-21 Multilingual Media Monitoring and Text Analysis – Challenges for Highly Inflected Languages....Pages 22-33 Spoken Dialogs with Children for Science Learning and Literacy....Pages 34-34 Statecharts and SCXML for Dialogue Management....Pages 35-35 A Comparison of Deep Neural Network Training Methods for Large Vocabulary Speech Recognition....Pages 36-43 A Comparison of Two Approaches to Bilingual HMM-Based Speech Synthesis....Pages 44-51 A Direct Criterion Minimization Based fMLLR via Gradient Descend....Pages 52-59 A Machine Learning Based Approach for Vocabulary Selection for Speech Transcription....Pages 60-67 A New State-of-The-Art Czech Named Entity Recognizer....Pages 68-75 Algorithms for Dysfluency Detection in Symbolic Sequences Using Suffix Arrays....Pages 76-83 Analysis and Combination of Forward and Backward Based Decoders for Improved Speech Transcription....Pages 84-91 Annotating Signs of Syntactic Complexity to Support Sentence Simplification....Pages 92-104 Application of LSTM Neural Networks in Language Modelling....Pages 105-112 Automatic Laughter Detection in Spontaneous Speech Using GMM–SVM Method....Pages 113-120 Automatic Machine Translation Evaluation with Part-of-Speech Information....Pages 121-128 Automatic Extraction of Polish Language Errors from Text Edition History....Pages 129-136 Bilingual Voice Conversion by Weighted Frequency Warping Based on Formant Space....Pages 137-144 Building a Hybrid: Chatterbot – Dialog System....Pages 145-152 CRF-Based Czech Named Entity Recognizer and Consolidation of Czech NER Research....Pages 153-160 Comparison and Analysis of Several Phonetic Decoding Approaches....Pages 161-168 Concatenation Artifact Detection Trained from Listeners Evaluations....Pages 169-176 Configuring TTS Evaluation Method Based on Unit Cost Outlier Detection....Pages 177-184 Development and Evaluation of Spoken Dialog Systems with One or Two Agents through Two Domains....Pages 185-192 Distant Supervision Learning of DBPedia Relations....Pages 193-200 Downdating Lexicon and Language Model for Automatic Transcription of Czech Historical Spoken Documents....Pages 201-208 Dynamic Threshold Selection Method for Multi-label Newspaper Topic Identification....Pages 209-216 Efficiency of Multi-tap Text Entry Method on Interactive Television....Pages 217-224 English Nominal Compound Detection with Wikipedia-Based Methods....Pages 225-232 Evaluating Voice Quality and Speech Synthesis Using Crowdsourcing....Pages 233-240 Experiment with Evaluation of Quality of the Synthetic Speech by the GMM Classifier....Pages 241-248 Experiments on Reducing Footprint of Unit Selection TTS System....Pages 249-256 Expressive Speech Synthesis for Urgent Warning Messages Generation in Romani and Slovak....Pages 257-264 Extracting Relations between Arabic Named Entities....Pages 265-271 Foot Detection in Czech Using Pitch Information and HMM....Pages 272-279 Improved Hungarian Morphological Disambiguation with Tagger Combination....Pages 280-287 Improving Dependency Parsing by Filtering Linguistic Noise....Pages 288-294 Improving Speech Recognition by Detecting Foreign Inclusions and Generating Pronunciations....Pages 295-302 Intensifying Verb Prefix Patterns in Czech and Russian....Pages 303-310 K-Component Adaptive Recurrent Neural Network Language Models....Pages 311-318 LIUM ASR System for ETAPE French Evaluation Campaign: Experiments on System Combination Using Open-Source Recognizers....Pages 319-326 Lexical Stress-Based Morphological Decomposition and Its Application for Ukrainian Speech Recognition....Pages 327-334 Motion Event in Croatian, English, German and Italian Concerning Path Prefixes and Prepositions....Pages 335-342 Multi-label Document Classification in Czech....Pages 343-351 On Behaviour of PLDA Models in the Task of Speaker Recognition....Pages 352-359 On the Quantitative and Qualitative Speech Changes of the Czech Radio Broadcasts News within Years 1969–2005....Pages 360-368 On the Use of Phoneme Lattices in Spoken Language Understanding....Pages 369-377 Online Speaker Adaptation of an Acoustic Model Using Face Recognition....Pages 378-385 Ontology of Rhetorical Figures for Serbian....Pages 386-393 Phoneme Set and Pronouncing Dictionary Creation for Large Vocabulary Continuous Speech Recognition of Vietnamese....Pages 394-401 Phonetic Spoken Term Detection in Large Audio Archive Using the WFST Framework....Pages 402-409 Ranking Machine Translation Systems via Post-editing....Pages 410-418 Recursive Part-of-Speech Tagging Using Word Structures....Pages 419-425 Resolving Ambiguities in Sentence Boundary Detection in Russian Spontaneous Speech....Pages 426-433 Revealing Prevailing Semantic Contents of Clusters Generated from Untagged Freely Written Text Documents in Natural Languages....Pages 434-441 Robust Methodology for TTS Enhancement Evaluation....Pages 442-449 Romanian Syllabication Using Machine Learning....Pages 450-456 SVM-Based Detection of Misannotated Words in Read Speech Corpora....Pages 457-464 Scratching the Surface of Possible Translations....Pages 465-474 Selecting and Weighting N-Grams to Identify 1100 Languages....Pages 475-483 Semantic Spaces for Sentiment Analysis....Pages 484-491 Semi-automatic Verb Valence Frame Assignment through VerbNet Classification....Pages 492-500 Speaker-Specific Pronunciation for Speech Synthesis....Pages 501-508 Structuring a Multimedia Tri-Dialectal Dictionary....Pages 509-518 Stylistic Changes for Temporal Text Classification....Pages 519-526 SummEC: A Summarization Engine for Czech....Pages 527-535 Text-to-Speech Alignment for Imperfect Transcriptions....Pages 536-543 The CNG Corpus of European Portuguese Children’s Speech....Pages 544-551 The Joint Optimization of Spectro-Temporal Features and Neural Net Classifiers....Pages 552-559 Three Syntactic Formalisms for Data-Driven Dependency Parsing of Croatian....Pages 560-567 Topic Models for Comparative Summarization....Pages 568-574 Using Low-Cost Annotation to Train a Reliable Czech Shallow Parser....Pages 575-582 Verb Subcategorisation Acquisition for Estonian Based on Morphological Information....Pages 583-590 Whispered Speech Database: Design, Processing and Application....Pages 591-598 Back Matter....Pages -