Blind Speech Separation

This is the world’s first edited book on independent component analysis (ICA)-based blind source separation (BSS) of convolutive mixtures of speech. This book brings together a small number of leading researchers to provide tutorial-like and in-depth treatment on major ICA-based BSS topics, with the objective of becoming the definitive source for current, comprehensive, authoritative, and yet accessible treatment.

Blind Source Separation intends to report the new results of the efforts on the study of Blind Source Separation (BSS). The book collects novel research ideas and some training in BSS, independent component analysis (ICA), artificial intelligence and signal processing applications. Furthermore, the research results previously scattered in many journals and conferences worldwide are methodically edited and presented in a unified form. The book is likely to be of interest to university researchers, R&D engineers and graduate students in computer science and electronics who wish to learn the core principles, methods, algorithms and applications of BSS. Dr. Ganesh R. Naik works at University of Technology, Sydney, Australia; Dr. Wenwu Wang works at University of Surrey, UK.

Edited by the people who were forerunners in creating the field, together with contributions from 34 leading international experts, this handbook provides the definitive reference on Blind Source Separation, giving a broad and comprehensive description of all the core principles and methods, numerical algorithms and major applications in the fields of telecommunications, biomedical engineering and audio, acoustic and speech processing. Going beyond a machine learning perspective, the book reflects recent results in signal processing and numerical analysis, and includes topics such as optimization criteria, mathematical tools, the design of numerical algorithms, convolutive mixtures, and time frequency approaches. This Handbook is an ideal reference for university researchers, R&D engineers and graduates wishing to learn the core principles, methods, algorithms, and applications of Blind Source Separation. Covers the principles and major techniques and methods in one book Edited by the pioneers in the field with contributions from 34 of the world’s experts Describes the main existing numerical algorithms and gives practical advice on their design Covers the latest cutting edge topics: second order methods; algebraic identification of under-determined mixtures, time-frequency methods, Bayesian approaches, blind identification under non negativity approaches, semi-blind methods for communications Shows the applications of the methods to key application areas such as telecommunications, biomedical engineering, speech, acoustic, audio and music processing, while also giving a general method for developing applications

This book is appropriate for those specializing in speech science, hearing science, neuroscience, or computer science and engineers working on applications such as automatic speech recognition, cochlear implants, hands-free telephones, sound recording, multimedia indexing and retrieval.

This book provides a detailed survey of the methods that were recently developed to handle advanced versions of the blind source separation problem, which involve several types of nonlinear mixtures. Another attractive feature of the book is that it is based on a coherent framework. More precisely, the authors first present a general procedure for developing blind source separation methods. Then, all reported methods are defined with respect to this procedure. This allows the reader not only to more easily follow the description of each method but also to see how these methods relate to one another. The coherence of this book also results from the fact that the same notations are used throughout the chapters for the quantities (source signals and so on) that are used in various methods. Finally, among the quite varied types of processing methods that are presented in this book, a significant part of this description is dedicated to methods based on artificial neural networks, especially recurrent ones, which are currently of high interest to the data analysis and machine learning community in general, beyond the more specific signal processing and blind source separation communities.

This book provides the first comprehensive overview of the fascinating topic of audio source separation based on non-negative matrix factorization, deep neural networks, and sparse component analysis. The first section of the book covers single channel source separation based on non-negative matrix factorization (NMF). After an introduction to the technique, two further chapters describe separation of known sources using non-negative spectrogram factorization, and temporal NMF models. In section two, NMF methods are extended to multi-channel source separation. Section three introduces deep neural network (DNN) techniques, with chapters on multichannel and single channel separation, and a further chapter on DNN based mask estimation for monaural speech separation. In section four, sparse component analysis (SCA) is discussed, with chapters on source separation using audio directional statistics modelling, multi-microphone MMSE-based techniques and diffusion map methods. The book brings together leading researchers to provide tutorial-like and in-depth treatments on major audio source separation topics, with the objective of becoming the definitive source for a comprehensive, authoritative, and accessible treatment. This book is written for graduate students and researchers who are interested in audio source separation techniques based on NMF, DNN and SCA.

A systematic exploration of both classic and contemporaryalgorithms in blind source separation with practical casestudies The book presents an overview of Blind Source Separation, arelatively new signal processing method. Due to themultidisciplinary nature of the subject, the book has been writtenso as to appeal to an audience from very different backgrounds.Basic mathematical skills (e.g. on matrix algebra and foundationsof probability theory) are essential in order to understand thealgorithms, although the book is written in an introductory,accessible style. This book offers a general overview of the basics of BlindSource Separation, important solutions and algorithms, and in-depthcoverage of applications in image feature extraction, remotesensing image fusion, mixed-pixel decomposition of SAR images,image object recognition fMRI medical image processing, geochemicaland geophysical data mining, mineral resources prediction andgeoanomalies information recognition. Firstly, the background andtheory basics of blind source separation are introduced, whichprovides the foundation for the following work. Matrix operation,foundations of probability theory and information theory basics areincluded here. There follows the fundamental mathematical model andfairly new but relatively established blind source separationalgorithms, such as Independent Component Analysis (ICA) and itsimproved algorithms (Fast ICA, Maximum Likelihood ICA, OvercompleteICA, Kernel ICA, Flexible ICA, Non-negative ICA, Constrained ICA,Optimised ICA). The last part of the book considers the very recentalgorithms in BSS e.g. Sparse Component Analysis (SCA) andNon-negative Matrix Factorization (NMF). Meanwhile, in-depth casesare presented for each algorithm in order to help the readerunderstand the algorithm and its application field. A systematic exploration of both classic and contemporaryalgorithms in blind source separation with practical casestudies Presents new improved algorithms aimed at differentapplications, such as image feature extraction, remote sensingimage fusion, mixed-pixel decomposition of SAR images, image objectrecognition, and MRI medical image processing With applications in geochemical and geophysical data mining,mineral resources prediction and geoanomalies informationrecognition Written by an expert team with accredited innovations in blindsource separation and its applications in natural science Accompanying website includes a software system providing codesfor most of the algorithms mentioned in the book, enhancing thelearning experience Essential reading for postgraduate students and researchersengaged in the area of signal processing, data mining, imageprocessing and recognition, information, geosciences, lifesciences.

"Blind Signal Processing: Theory and Practice" not only introduces related fundamental mathematics, but also reflects the numerous advances in the field, such as probability density estimation-based processing algorithms, underdetermined models, complex value methods, uncertainty of order in the separation of convolutive mixtures in frequency domains, and feature extraction using Independent Component Analysis (ICA). At the end of the book, results from a study conducted at Shanghai Jiao Tong University in the areas of speech signal processing, underwater signals, image feature extraction, data compression, and the like are discussed. This book will be of particular interest to advanced undergraduate students, graduate students, university instructors and research scientists in related disciplines. Xizhi Shi is a Professor at Shanghai Jiao Tong University.