Speech Recognition and Understanding: Recent Advances, Trends and Applications (Nato ASI Subseries F: (75))

The book collects the contributions to the NATO Advanced Study Institute on "Speech Recognition and Understanding: Recent Advances, Trends and Applications", held in Cetraro, Italy, during the first two weeks of July 1990. This Institute focused on three topics that are considered of particular interest and rich of i'p.novation by researchers in the fields of speech recognition and understanding: Advances in Hidden Markov modeling, connectionist approaches to speech and language modeling, and linguistic processing including language and dialogue modeling. The purpose of any ASI is that of encouraging scientific communications between researchers of NATO countries through advanced tutorials and presentations: excellent tutorials were offered by invited speakers that present in this book 15 papers which sum­ marize or detail the topics covered in their lectures. The lectures were complemented by discussions, panel sections and by the presentation of related works carried on by some of the attending researchers: these presentations have been collected in 42 short contributions to the Proceedings. This volume, that the reader can find useful for an overview, although incomplete, of the state of the art in speech understanding, is divided into 6 Parts. Front Matter....Pages I-XI Front Matter....Pages 1-1 Hidden Markov Models for Speech Recognition — Strengths and Limitations....Pages 3-29 Hidden Markov Models and Speaker Adaptation....Pages 31-57 A 20,000 word automatic speech recognizer....Pages 59-64 Automatic adjustments of the Markov models topology for speech recognition applications over the telephone....Pages 65-70 Phonetic Structure Inference of Phonemic HMM....Pages 71-76 Phonetic Units and Phonotactical Structure Inference by Ergodic Hidden Markov Models....Pages 77-82 Clustering of Gaussian densities in hidden Markov models....Pages 83-88 Developments in High-Performance Connected Digit Recognition....Pages 89-94 Robust Speaker-Independent Hidden Markov Model Based Word Spotter....Pages 95-100 Robust Speech Recognition in Noisy and Reverberant Environments....Pages 101-106 An ISDN speech server based on speaker independent continuous Hidden Markov Models....Pages 107-112 RAMSES: A Spanish Demisyllable Based Continuous Speech Recognition System....Pages 113-118 Speaker Independent 1000 Words Speech Recognition in Spanish....Pages 119-124 Continuously Variable Transition Probability HMM for Speech Recognition....Pages 125-130 Front Matter....Pages 131-131 Context-Dependent Phonetic Hidden Markov Models for Speaker-Independent Continuous Speech Recognition....Pages 133-133 Speaker Independent Continuous Speech Recognition Using Continuous Density Hidden Markov Models....Pages 135-163 A Fast Lexical Selection Strategy for Large Vocabulary Continuous Speech Recognition....Pages 165-170 Performance of a Speaker-Independent Continuous Speech Recognizer....Pages 171-179 Automatic Transformation of Speech Databases for Continuous Speech Recognition....Pages 181-186 Iterative Optimization of the Data Driven Analysis in Continuous Speech....Pages 187-192 Front Matter....Pages 131-131 Syllable-based stochastic models for continuous speech recognition....Pages 193-198 Word Hypothesization in Continuous Speech Recognition....Pages 199-203 Phone Recognition Using High Order Phonotactic Constraints....Pages 205-210 An Efficient Structure for Continuous Speech Recognition....Pages 211-216 Search Organization for Large Vocabulary Continuous Speech Recognition....Pages 217-222 Front Matter....Pages 223-223 Neural Networks or Hidden Markov Models for Automatic Speech Recognition: Is there a Choice?....Pages 225-236 Neural Networks for Continuous Speech Recognition....Pages 237-257 Connectionist Large Vocabulary Speech Recognition....Pages 259-273 The cortical column as a model for speech recognition: principles and first experiments....Pages 275-292 Radial Basis Functions for Speech Recognition....Pages 293-298 Phonetic features extraction using Time-Delay Neural Networks....Pages 299-304 Improved Broad Phonetic Classification and Segmentation with an Auditory Model....Pages 305-310 Automatic Learning of a Production Rule System for Acoustic-Phonetic Decoding....Pages 311-316 Front Matter....Pages 317-317 Stochastic Grammars and Pattern Recognition....Pages 319-344 Basic Methods of Probabilistic Context Free Grammars....Pages 345-360 A Probabilistic Approach to Person-Robot Dialogue....Pages 361-380 Experimenting Text Creation by Natural-Language, Large-Vocabulary Speech Recognition....Pages 381-387 DUALGRAM: An Efficient Method for Representing Limited-Domain Language Models....Pages 389-394 Strategies for Speech Recognition and Understanding using Layered Protocols....Pages 395-400 Front Matter....Pages 401-401 TINA: A Probabilistic Syntactic Parser for Speech Understanding Systems....Pages 403-414 Front Matter....Pages 401-401 The Voyager Speech Understanding System: A Progress Report....Pages 415-424 The Interaction of Word Recognition and Linguistic Processing in Speech Understanding....Pages 425-453 Linguistic Processing in a Speech Understanding System....Pages 455-479 Linguistic Tools for Speech Recognition and Understanding....Pages 481-485 Evidential reasoning and the combination of knowledge and statistical techniques in syllable based speech recognition....Pages 487-492 Front Matter....Pages 493-493 Data Base Management for Use with Acoustic-Phonetic Speech Data Bases....Pages 495-500 BPF Outputs Compared with Formant Frequencies and LPC’s for the Recognition of Vowels....Pages 501-506 A Codification of Error Signal by Splines Functions....Pages 507-512 Specific Distance for Feature Selection in Speech Recognition....Pages 513-518 Multiple Template Modeling of Sublexical Units....Pages 519-524 Learning Structural Models of Sublexical Units....Pages 525-530 On the use of Negative Samples in the MGGI Methodology and its application for Difficult Vocabulary Recognition Tasks....Pages 531-536 A New Method for Dynamic Time Alignment of Speech Waveforms....Pages 537-542 A New Technique for Automatic Segmentation of Continuous Speech....Pages 543-548 Segmentation of speech based upon a linear model of the effects of coarticulation....Pages 549-554 Back Matter....Pages 555-564 This book collects the contribution to a NATO seminar on speech recognition and understanding which focused on advan- ces in Hidden Markov Modeling, connectionist approaches to speech and language modeling, and linguistic processing in- cluding language and dialogue modeling. Proceedings of the NATO Advanced "Study Institute on Speech Recognition and Understanding. Recent Advances, Trends and Applications" held in Cetraro, Italy, July 1-13, 1990