Remarkable progress is being made in spoken language processing, but many powerful techniques have remained hidden in conference proceedings and academic papers, inaccessible to most practitioners. In this book, the leaders of the Speech Technology Group at Microsoft Research share these advances -- presenting not just the latest theory, but practical techniques for building commercially viable products.KEY TOPICS: Spoken Language Processing draws upon the latest advances and techniques from multiple fields: acoustics, phonology, phonetics, linguistics, semantics, pragmatics, computer science, electrical engineering, mathematics, syntax, psychology, and beyond. The book begins by presenting essential background on speech production and perception, probability and information theory, and pattern recognition. The authors demonstrate how to extract useful information from the speech signal; then present a variety of contemporary speech recognition techniques, including hidden Markov models, acoustic and language modeling, and techniques for improving resistance to environmental noise. Coverage includes decoders, search algorithms, large vocabulary speech recognition techniques, text-to-speech, spoken language dialog management, user interfaces, and interaction with non-speech interface modalities. The authors also present detailed case studies based on Microsoft's advanced prototypes, including the Whisper speech recognizer, Whistler text-to-speech system, and MiPad handheld computer.MARKET: For anyone involved with planning, designing, building, or purchasing spoken language technology.

After decades of research activity, Chinese spoken language processing (CSLP) has advanced considerably both in practical technology and theoretical discovery. In this book, the editors provide both an introduction to the field as well as unique research problems with their solutions in various areas of CSLP. The contributions represent pioneering efforts ranging from CSLP principles to technologies and applications, with each chapter encapsulating a single problem and its solutions.A commemorative volume for the 10th anniversary of the international symposium on CSLP in Singapore, this is a valuable reference for established researchers and an excellent introduction for those interested in the area of CSLP.

This book describes the Spoken Language Translator (SLT), one of the first major projects in the area of automatic speech translation.

Spoken language understanding (SLU) is an emerging field in between speech and language processing, investigating human/ machine and human/ human communication by leveraging technologies from signal processing, pattern recognition, machine learning and artificial intelligence. SLU systems are designed to extract the meaning from speech utterances and its applications are vast, from voice search in mobile devices to meeting summarization, attracting interest from both commercial and academic sectors. Both human/machine and human/human communications can benefit from the application of SLU, using differing tasks and approaches to better understand and utilize such communications. This book covers the state-of-the-art approaches for the most popular SLU tasks with chapters written by well-known researchers in the respective fields. Key features include: Presents a fully integrated view of the two distinct disciplines of speech processing and language processing for SLU tasks. Defines what is possible today for SLU as an enabling technology for enterprise (e.g., customer care centers or company meetings), and consumer (e.g., entertainment, mobile, car, robot, or smart environments) applications and outlines the key research areas. Provides a unique source of distilled information on methods for computer modeling of semantic information in human/machine and human/human conversations. This book can be successfully used for graduate courses in electronics engineering, computer science or computational linguistics. Moreover, technologists interested in processing spoken communications will find it a useful source of collated information of the topic drawn from the two distinct disciplines of speech processing and language processing under the new area of SLU.

Comprehensive textbook examining how both 'normal' and brain-damaged speakers process language in the brain.

An argument that the way we listen to speech is shaped by our experience with our native language. Understanding speech in our native tongue seems natural and effortless; listening to speech in a nonnative language is a different experience. In this book, Anne Cutler argues that listening to speech is a process of native listening because so much of it is exquisitely tailored to the requirements of the native language. Her cross-linguistic study (drawing on experimental work in languages that range from English and Dutch to Chinese and Japanese) documents what is universal and what is language specific in the way we listen to spoken language. Cutler describes the formidable range of mental tasks we carry out, all at once, with astonishing speed and accuracy, when we listen. These include evaluating probabilities arising from the structure of the native vocabulary, tracking information to locate the boundaries between words, paying attention to the way the words are pronounced, and assessing not only the sounds of speech but prosodic information that spans sequences of sounds. She describes infant speech perception, the consequences of language-specific specialization for listening to other languages, the flexibility and adaptability of listening (to our native languages), and how language-specificity and universality fit together in our language processing system. Drawing on her four decades of work as a psycholinguist, Cutler documents the recent growth in our knowledge about how spoken-word recognition works and the role of language structure in this process. Her book is a significant contribution to a vibrant and rapidly developing field.

Spoken Language Comprehensionis the first coherent presentation of an original detailed experimental and theoretical account of what are rationally taken to be "online" processing deficits that lie at the core of aphasic miscomprehension. It presents exciting work that is highly relevant to the important current debate about the nature of aphasic comprehension impairment and its relationship to models of normal functioning. Lorraine K. Tyler focuses on a crucial but neglected aspect of language disorders: how the real-time analysis processes involved in comprehending spoken language break down in acquired aphasia. She describes a new approach to the study of language disorders that specifies the processes involved in the immediate construction of various types of linguistic representations. Her unique large-scale analysis makes possible the evaluation of various theoretical accounts of the underlying basis of different kinds of aphasic deficits. By developing a set of experimental tests designed to detect specific deficits in the principal categories of real-time comprehension, Tyler constructs a processing profile of ten patients that shows where each patient performs normally and where performance breaks down. This provides a detailed picture of a patient's ability to perform the appropriate analyses of speech input: breaking down the speech signal, recognizing words, making the appropriate form-function mapping, and constructing the appropriate types of higher-level representations (syntactic, semantic, pragmatic, and prosodic). Data from standard tests of comprehension deficits are also included, which permits comparison of performance in various tasks and among patients to see where differences and similarities emerge. Lorraine Komisarjevsky Tyler is Professor of Psychology at the University of London.

This volume brings together papers from the areas of psychology, general linguistics, psycholinguistics, as well as from simultaneous interpreting. Their common focus is how theories and methodologies from various disciplines can be applied to the study of simultaneous interpreting, and also to suggest ways in which the study of simultaneous interpreting in its turn might contribute to knowledge in other areas. General topics dealt with include memory, language processing, bilingual processing, and second language acquisition. The articles more specifically focused on simultaneous interpreting discuss implications of the general topics and report on empirical studies on expertise in interpreting and on phonological interference in spoken language interpreting. Requirements for further interdisciplinary research in the context of simultaneous interpreting are considered. There is also a discussion of transcription conventions for simultaneous interpreting.

Speech and language technologies continue to grow in importance as they are used to create natural and efficient interfaces between people and machines, and to automatically transcribe, extract, analyze, and route information from high-volume streams of spoken and written information. The workshops on Mathematical Foundations of Speech Processing and Natural Language Modeling were held in the Fall of 2000 at the University of Minnesota's NSF-sponsored Institute for Mathematics and Its Applications, as part of a "Mathematics in Multimedia" year-long program. Each workshop brought together researchers in the respective technologies on the one hand, and mathematicians and statisticians on the other hand, for an intensive week of cross-fertilization. There is a long history of benefit from introducing mathematical techniques and ideas to speech and language technologies. Examples include the source-channel paradigm, hidden Markov models, decision trees, exponential models and formal languages theory. It is likely that new mathematical techniques, or novel applications of existing techniques, will once again prove pivotal for moving the field forward. This volume consists of original contributions presented by participants during the two workshops. Topics include language modeling, prosody, acoustic-phonetic modeling, and statistical methodology.