Speech Recognition and Understanding

This book entitled, "Introduction to Speech and Language Therapy", has been designed to add to the knowledge of researchers, scholars and the students of second language learners to enlighten them with various aspects of the speech, the language and the methods and techniques of its acquisition. It takes the readers through an overview of speech and language therapy and how these concepts assist the children and the adults in gaining the intricate nuances of the language. It also talks about the role and technique of speech and language proper and its assessment in the teaching of second language. Additionally, the book sheds light on the linguistic theories of speech, the teaching of language for children and the techniques of vocabulary development. Automatic Speech Recognition (ASR) is one of the greatest technical challenges of modern times and has been attracting the attention of researchers around the world for more than half a century. As with all speech technologies, this is a multidisciplinary problem that requires knowledge in many areas, from acoustics, phonetics and linguistics, to mathematics, telecommunications, signal processing and programming. A special problem is the fact that it is a problem that is extremely language-dependent. Speech recognition or speech to text includes capturing and digitizing the sound waves, transformation of basic linguistic units or phonemes, constructing words from phonemes and contextually analyzing the words to ensure the correct spelling of words that sounds the same. Approach: Studying the possibility of designing a software system using one of the techniques of Artificial Intelligence (AI) applications neuron networks where this system is able to distinguish the sound signals and neural networks of irregular users. Fixed weights are trained on those forms first and then the system gives the output match for each of these formats and high speed. The proposed neural network study is based on solutions of speech recognition tasks, detecting signals using angular modulation and detection of modulated techniques. Cover 1 Title Page 5 Copyright 6 DECLARATION 7 ABOUT THE EDITOR 9 TABLE OF CONTENTS 11 List of Contributors 17 List of Abbreviations 23 Preface 27 Section 1: Methods and Approaches for Speech Recognition 29 Chapter 1 Artificial Intelligence for Speech Recognition Based on Neural Networks 31 Abstract 31 Introduction 32 Pattern Recognition 33 Neural Networks 34 Procedure Works 35 Conclusion 41 References 42 Chapter 2 An HMM-Like Dynamic Time Warping Scheme for Automatic Speech Recognition 43 Abstract 43 Introduction 44 Speech Recognition By DTW 47 The Proposed Hmm-Like DTW Approach for Speech Recognition 49 Hmm-Like DTW 51 Experiments And Results 56 Conclusions 59 References 60 Chapter 3 Direct Recovery of Clean Speech Using a Hybrid Noise Suppression Algorithm for Robust Speech Recognition System 63 Abstract 63 Introduction 64 System Model 66 Algorithm Description 67 Algorithm Evaluation 76 Conclusion 81 Supplementary Materials 82 References 84 Chapter 4 Deep Neural Learning Adaptive Sequential Monte Carlo for Automatic Image and Speech Recognition 87 Abstract 88 Introduction 88 Materials and Methods 90 Results and Discussion 95 Conclusions 103 Acknowledgments 104 References 105 Chapter 5 A Fast Learning Method for Multilayer Perceptrons in Automatic Speech Recognition Systems 109 Abstract 110 Introduction 110 A Fast Learning Method 112 Experiments 116 Conclusions 123 Acknowledgments 123 References 124 Section 2: Speech Recognition for Different Languages 129 Chapter 6 Development of Application Specific Continuous Speech Recognition System in Hindi 131 Abstract 131 Introduction 132 Automatic Speech Recognition System 134 The Training Methodology 139 Evaluation Methodology 143 Results And Discussion 146 Conclusion And Future Work 146 References 148 Chapter 7 Multitask Learning with Local Attention for Tibetan Speech Recognition 151 Abstract 151 Introduction 152 Related Work 153 Methods 154 Experiments 159 Conclusions 167 Authors’ Contributions 167 Acknowledgments 167 References 168 Chapter 8 Arabic Speech Recognition System Based on MFCC and HMMs 171 Introduction 172 Mel Frequency Cepstral Coefficients (Mfcc) 173 Hidden Markov Model (Hmm) 174 Experimental Results 176 Conclusion 178 References 179 Chapter 9 Using Morphological Data in Language Modeling for Serbian Large Vocabulary Speech Recognition 181 Abstract 181 Introduction 182 Relevant Previous Work 184 Materials And Methods 185 Results And Discussion 192 Conclusions 196 Data Availability 196 Acknowledgments 197 References 198 Chapter 10 Phoneme Sequence Modeling in the Context of Speech Signal Recognition in Language “Baoule” 203 Abstract 203 Introduction 204 The Speech Signals 205 System Overview 206 Hidden Markov Model Discrete Time 209 Implementation 220 Conclusions 226 Annexes 227 References 230 Section 3: Applications with Speech Recognition 231 Chapter 11 An Overview of Basics Speech Recognition and Autonomous Approach for Smart Home IOT Low Power Devices 233 Abstract 233 Introduction 234 Overview State of the Art 234 Our Methodology 240 Algorithm Description 243 Recognition Technique 246 Results 248 Conclusion 254 References 255 Chapter 12 BigEar: Ubiquitous Wireless Low-Budget Speech Capturing Interface 257 Abstract 257 Introduction 258 Related Work 260 Bigear Architecture 263 Speech Acquisition Model and Implementation 267 Speech Reconstruction 271 Bigear Simulation and Model Validation 278 Experimental Results and Evaluation 281 Conclusions and Future Works 285 References 287 Chapter 13 Using Speech Recognition in Learning Primary School Mathematics via Explain, Instruct and Facilitate Techniques 289 Abstract 289 Introduction 290 Materials and Methods 293 Results and Discussions 299 Conclusions and Recommendations 320 Acknowledgements 321 References 322 Chapter 14 A Prototype of a Semantic Platform with a Speech Recognition System for Visual Impaired People 325 Abstract 325 Introduction 326 Review of Literature 327 Current Problems of Web Platforms for Accessibility 328 Prototype of Semantic Platform with Speech Recognition System 329 Conceptual Scheme of Architecture 331 Expected Contributions and Future Work 333 References 334 Section 4: Language Understanding Technology 337 Chapter 15 English Sentence Recognition Based on HMM and Clustering 339 Abstract 339 Introduction 340 Whole Design Process 341 Core Algorithm 342 Experimental Results and Analysis 345 Conclusion 347 Acknowledgements 348 References 349 Chapter 16 A Comparative Study to Understanding about Poetics Based on Natural Language Processing 351 Abstract 351 Introduction 352 Materials and Method 353 Results 357 Discussion 358 Conclusion 360 References 361 Chapter 17 Semi-Supervised Learning of Statistical Models for Natural Language Understanding 363 Abstract 363 Introduction 364 Related Work 366 The Proposed Framework 369 Experimental Results 376 Conclusions 384 Acknowledgments 385 References 386 Chapter 18 Linguistic Factors Influencing Speech Audiometric Assessment 389 Abstract 389 Introduction 390 Linguistic Cues to Speech Understanding 391 Aim And Research Questions 392 Syntactic Complexity, Cognitive Load, and Speech Understanding 393 The Role of Open Versus Closed Word Classes In Sentence Understanding 396 Materials and Method 398 Results 403 Discussion 411 Conclusion 415 Acknowledgments 416 References 417 Index 423 Back Cover 428 Covers various aspects of speech, language and the methods and techniques of its acquisition. The book takes readers through an overview of speech and language therapy and how these concepts assist children and the adults in gaining the intricate nuances of language.