You are here

FAU Collections » FAU Research Repository » FAU Theses and Dissertations

Sensitivity analysis of blind separation of speech mixtures

Download pdf | Full Screen View

Date Issued:
2010
Summary:
Blind source separation (BSS) refers to a class of methods by which multiple sensor signals are combined with the aim of estimating the original source signals. Independent component analysis (ICA) is one such method that effectively resolves static linear combinations of independent non-Gaussian distributions. We propose a method that can track variations in the mixing system by seeking a compromise between adaptive and block methods by using mini-batches. The resulting permutation indeterminacy is resolved based on the correlation continuity principle. Methods employing higher order cumulants in the separation criterion are susceptible to outliers in the finite sample case. We propose a robust method based on low-order non-integer moments by exploiting the Laplacian model of speech signals. We study separation methods for even (over)-determined linear convolutive mixtures in the frequency domain based on joint diagonalization of matrices employing time-varying second order statistics. We investigate the sources affecting the sensitivity of the solution under the finite sample case such as the set size, overlap amount and cross-spectrum estimation methods.
Title: Sensitivity analysis of blind separation of speech mixtures.
81 views
14 downloads
Name(s): Bulek, Savaskan.
College of Engineering and Computer Science
Department of Computer and Electrical Engineering and Computer Science
Type of Resource: text
Genre: Electronic Thesis Or Dissertation
Date Issued: 2010
Publisher: Florida Atlantic University
Physical Form: electronic
Extent: xi, 112 p. : ill. (some col.)
Language(s): English
Summary: Blind source separation (BSS) refers to a class of methods by which multiple sensor signals are combined with the aim of estimating the original source signals. Independent component analysis (ICA) is one such method that effectively resolves static linear combinations of independent non-Gaussian distributions. We propose a method that can track variations in the mixing system by seeking a compromise between adaptive and block methods by using mini-batches. The resulting permutation indeterminacy is resolved based on the correlation continuity principle. Methods employing higher order cumulants in the separation criterion are susceptible to outliers in the finite sample case. We propose a robust method based on low-order non-integer moments by exploiting the Laplacian model of speech signals. We study separation methods for even (over)-determined linear convolutive mixtures in the frequency domain based on joint diagonalization of matrices employing time-varying second order statistics. We investigate the sources affecting the sensitivity of the solution under the finite sample case such as the set size, overlap amount and cross-spectrum estimation methods.
Identifier: 700217627 (oclc), 2953201 (digitool), FADT2953201 (IID), fau:3557 (fedora)
Note(s): by Savaskan Bulek.
Thesis (Ph.D.)--Florida Atlantic University, 2010.
Includes bibliography.
Electronic reproduction. Boca Raton, Fla., 2010. Mode of access: World Wide Web.
Subject(s): Blind source separation -- Mathematical models
Signal processing -- Digital techniques
Neural networks (Computer science)
Automatic speech recognition
Speech processing systems
Persistent Link to This Record: http://purl.flvc.org/FAU/2953201
Use and Reproduction: http://rightsstatements.org/vocab/InC/1.0/
Host Institution: FAU