Audio Source Separation and Speech Enhancement

Most books focus on one of the three historical paths: sensor array processing, computational auditory scene analysis, or independent component analysis. This book will be the first to provide a comprehensive overview of the field by presenting the common foundations and the differences between these techniques in a structured way. It provides both introductory and advanced material enabling readers to better grasp the field and acquire cross-domain knowledge, and offers detailed descriptions of certain core signal processing algorithms that can be used to separate signals, and their implementation. The book will cover time-frequency processing, acoustics and multichannel source activity detection, localization and tracking; single-channel enhancement and separation, multichannel enhancement and separation, and application scenarios and perspectives.