This paper places the perception of noise annoyance within the framework of environmental assessment. It is shown that due to the inherent vagueness and uncertainties in the involved concepts, available data and knowledge, fuzzy techniques are very well suited for the modeling of this perception. This idea is then used to construct a fuzzy rule based model for the prediction of the level of general noise annoyance. It turns out that this model can be decomposed in building blocks, each describing the noise annoyance caused by an individual source. Finally, the models are tested against data collected in social surveys and the results are reported.