Nonlinear speech enhancement: An overview

Citation

Hussain A, Chetouani M, Squartini S, Bastari A & Piazza F (2007) Nonlinear speech enhancement: An overview. In: Stylianou Y, Faundez-Zanuy M & Esposito A (eds.) Progress in Nonlinear Speech Processing. Lecture Notes in Computer Science, 4391. Berlin Heidelberg: Springer, pp. 217-248. http://link.springer.com/chapter/10.1007/978-3-540-71505-4_12#; https://doi.org/10.1007/978-3-540-71505-4_12

Abstract
This paper deals with the problem of enhancing the quality of speechsignals, which has received growing attention in the last few decades. Many differentapproaches have been proposed in the literature under various configurationsand operating hypotheses. The aim of this paper is to give an overview ofthe main classes of noise reduction algorithms proposed to-date, focusing on thecase of additive independent noise. In this context, we first distinguish betweensingle and multi channel solutions, with the former generally shown to be basedon statistical estimation of the involved signals whereas the latter usually employadaptive procedures (as in the classical adaptive noise cancellationscheme). Within these two general classes, we distinguish between certain subfamiliesof algorithms. Subsequently, the impact of nonlinearity on the speechenhancement problem is highlighted: the lack of perfect linearity in relatedprocesses and the non-Gaussian nature of the involved signals are shown tohave motivated several researchers to propose a range of efficient nonlineartechniques for speech enhancement. Finally, the paper summarizes (in tabularform) for comparative purposes, the general features, list of operating assumptions,the relative advantages and drawbacks, and the various types of nonlineartechniques for each class of speech enhancement strategy.

Keywords
advantage; algorithm; Algorithms; Attention; C; class; context; DECADE; enhancement; Estimation; Feature; features; IMPACT; LITERATURE; QUALITY; RANGE; reduction; researchers; SINGLE; Speech; Techniques