| Fundamentals of Speaker Recognition | |
| Homayoon Beigi |
|
From the Preface: […] In this book I have tried to cover as much detail as possible to keep most of the necessary information self-contained and rigorous. Although, you will see many references presented at the end of each chapter and finally as a collection in a full bibliography […] To be able to present the details, and yet have a smooth narrative in the main text, a large amount of detailed material is included in the last 4 chapters of the book, categorized as Background Material […] The main narrative of the book has three major parts: Part I covers the introductory and basic theory of the subject including anatomy, signal representation, phonetics, signal processing and feature extraction, probability theory, information theory, metrics and distortion measures, Bastian learning theory, parameter estimation and leaning, clustering, parameter transformation, hidden Markov modeling, neural networks, and support vector machines. The second part, advanced theory, covers subjects which deal more directly with speaker recognition. These topics are speaker modeling, speaker recognition implementation, and signal enhancement and compensation. Part III, practice, discusses topics specifically related to the implementation of speaker recognition or related issues. These are representation of results, time-lapse effects, adaptation techniques, and finally, overall design issues […] |
|