Since audio quality is an area of growing concern for many users of audio equipment, so is the area of audio quality assessment for its manufacturers. This assessment can be done by listening tests, however, there is a need for a cheaper, yet equally accurate method. Most of the objective computational models, which predict audio quality, concentrate only on the monophonic perception, which might lead to underestimation of spatial degradations
to audio. The aim of the project was to develop - based on perceptual models which are already available - a binaural model of auditory perception, which can be used to assess audio quality degradation, in both its spatial, and non-spatial character. One monophonic model (CASP) and three different binaural processors were considered. A listening test was conducted, in order
to validate the combined objective models, as well as to adjust some of their parameters and optimise their predictions. Finally, a combination of models and parameters, which seemed optimal, was chosen. However, if a truly optimal and
easy to use model should be developed, there are still some areas which need further investigation.