Perceptual speech quality assessment - a review
Rix, A.W.
Acoustics, Speech, and Signal Processing, 2004. Proceedings. (ICASSP apos;04). IEEE International Conference on
Volume 3, Issue , 17-21 May 2004 Page(s): iii - 1056-9 vol.3
Digital Object Identifier 10.1109/ICASSP.2004.1326730
Summary: This paper reviews the development of perceptually-motivated models for quality assessment of speech transmission/storage systems. The aim is to predict a subjective mean opinion score (MOS) for non-linear, time-variant distortions such as lossy coders, channel errors or noise reduction, particularly for telecommunications applications. Because linear methods have proven unsuitable for this purpose, many researchers have studied perceptual quality assessment using a comparison of auditory transforms to estimate quality. This work has led to several ITU standards. Non-intrusive models, arguably more suited to network monitoring, are the focus of much current interest. Intrusive, signal-based non-intrusive, and parametric non-intrusive models are discussed.
View citation and abstract |