Subjective evaluation and analysis of auditory attributes for underwater noise

Dimensionality and the corresponding physical characteristics of auditory attribute space for equal-loudness underwater noise are investigated. Firstly, cluster analysis and questionnaire surveys are performed to obtain a number of auditory attribute descriptors in Chinese language, and then based on paired comparison and semantic differential approach, a dissimilarity matrix and a subjective evaluation value of each descriptor are obtained. Finally, by use of multidimensional scaling and principal component analysis, auditory attribute space of underwater noise with equal-loudness and equal-duration is determined and then the five-dimensional space is composed by dark-roughness, fluctuation, fast-variety, sharpness and regularity.