No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

Single channel blind source separation (SCBSS) is an intensively researched field with numerous important applications. This research sets out to investigate the separation of monaural mixed audio recordings without relying on training knowledge. This research proposes a novel method based on variable regularised sparse nonnegative matrix factorization which decomposes an information-bearing matrix into two-dimensional convolution of factor matrices that represent the spectral basis and temporal code of the sources. In this work, a variational Bayesian approach has been developed for computing the sparsity parameters of the matrix factorization. To further improve the previous work, this research proposes a new method based on decomposing the mixture into a series of oscillatory components termed as the intrinsic mode functions (IMF). It is shown that IMFs have several desirable properties unique to SCBSS problem and how these properties can be advantaged to relax the constraints posed by the problem. In addition, this research develops a novel method for feature extraction using psycho-acoustic model. The monaural mixed signal is transformed to a cochleagram using the gammatone filterbank, whose bandwidths increase incrementally as the center frequency increases; thus resulting to non-uniform time-frequency (TF) resolution in the analysis of audio signal. Within this domain, a family of Itakura-Saito (IS) divergence based novel two-dimensional matrix factorization has been developed. The proposed matrix factorizations have the property of scale invariant which enables lower energy components in the cochleagram to be treated with equal importance as the high energy ones. Results show that all the developed algorithms presented in this thesis have outperformed conventional methods.

This thesis documents research in the field of sound diffusion for the live performance of electroacoustic music. Broad and inclusive ways of conceptual ising electroacoustic music are presented, with the intention of promoting the design of improved sound diffusion systems in the future. Having defined 'electroacoustic music' in terms of the technologies involved and the unique ways in which these creative frameworks are appropriated by practitioners (Chapter 1), a binary interpretation of the electroacoustic idiom, whereby musical philosophies can be regarded as either top-down or bottom-up, is given (Chapter 2). Discussion of the process of sound diffusion itself reveals two distinct performance praxes, which can also be characterised as top-down and bottom-up (Chapter 3). These differing ideologies, in addition to the technical demands of the electroacoustic idiom and the logistical demands of sound diffusion itself, must be accommodated by the sound diffusion system if live performances are to achieve the desired musical communication. It is argued that this is not presently the case. A system of criteria for the evaluation of sound diffusion systems is presented (Chapter 4). Two original concepts - the coherent audio source set (CASS) and coherent loudspeaker set (CLS) - are also presented; these are intended to be practically and theoretically useful in the field of sound diffusion. Several existing diffusion systems are evaluated in terms of these criteria (also Chapter 4). A description and evaluation of the M2 Sound Diffusion System, which was co-developed by the author as part of this research, is also given (Chapter 5). The final chapter describes ways in which superior future systems can be devised. These range from specific practical suggestions to general methodological recommendations. Overall, the intention is to provide an interpretation of the electroacoustic idiom that can be used as a heuristic tool 111 the design of new sound diffusion systems.

No description available.

No description available.