Fundamentals of Speaker Recognition

ISBN-13: 9780387775913 / Angielski / Twarda / 2011 / 942 str.

Beigi

Opis

Speaker Recognition is an emerging technology which has become quite popular due to its practicality of implementation in an existing telephone infrastructure, and the capability of remote biometric authentication. It is a multidisciplinary subject with information scattered throughout different research publications and textbooks. Speaker Recognition is getting more and more known for providing voice authentication over the telephones for helpdesks, call centers and other enterprise business for business process automation. Technology is evolving to support the consumer end solutions, where people have to understand the fundamentals of speaker recognition. A single textbook which treats Speaker Recognition with all the theoretical details is highly desirable by advanced level students in computer science and engineering, concentrating on biometrics, speech recognition, pattern recognition, signal processing and specifically speaker recognition. It will also be desired by developers of commercial technology and speech scientists. The Fundamentals of Speaker Recognition introduces the different branches of Speaker Recognition: Speaker Identification, Speaker Verification, Speaker Classification (Audio Event Classification), Speaker Detection, Speaker Tracking and more. The technical problems are formed and rigorously defined in detail, and practical algorithms are covered in this text. A complete picture is made of the relevance of the discussed algorithms and their usage in building a comprehensive Speaker Recognition System. This text book also includes examples and exercises at the end of each chapter (with a direct link to a solution's manual). The Fundamentals of Speaker Recognition is truly designed as a textbook for advanced level students in computer science and engineering. This textbook is also suitable for practitioners and researchers in industry.

Recenzje

From the reviews:

"... there is no comprehensive overview of speaker recognition.

Aside from the absence of a complete overview of the technology there is a commercial need for a good reference. The reason is that the commercial market for speaker recognition is growing and a need to have scientists and technology developers who understand and can work with this technology.

It goes into far more technical depth than other publications - mostly because those other publications cover other topics as well.

It is a scholarly book that does not appear to be representing knowledge about the topic through the filter of a specific product or kind of application. That is very good for scientists and scholars. Consequently, it would be a good text book or technical reference."

Judith Markowitz (J. Markowitz Consultants, Chicago, Illinois, USA)

"In short, this publication would be a service to many researchers and practitioners of this science and technology. I do not know of any other book on the subject. I am much impressed with the depth and breadth that the subject is going to be treated as evidenced from its table of contents."

Fereydoun Maali, Ph.D. (ICx Vision Systems, Phoenix, Arizona, USA)

"This book is unique in several respects. It is the most exhaustive text on speaker recognition available. It is also truly self-contained. About a third of the text is devoted to the background information needed for understanding speaker recognition technology. ... this text sets the standard for speaker recognition textbooks. I recommend it both as an introduction to the subject and as a reference on speaker recognition techniques." (Vladimir Botchev, ACM Computing Reviews, January, 2013)

Spis treści

Preface.- Basic Theory.- Introduction.- Speaker and Vocal Tract Modeling.- Signal Processing and Feature Extraction Techniques.- Data Representation and Probability Distributions.- Information Theory.- Metrics and Distortion Measures Bayesian Learning and Gaussian Mixture Modeling.- Parameter Estimation and Learning.- Hidden Markov Modeling (HMM).- Support Vector Machines.- Neural Networks.- Advanced Theory.- Speaker Modeling.- Algorithms.- Practice.- Speaker Recognition.- Overall Design.- Representation of Results.- Glossary.- Index.