Probability distribution estimation of music signals in time and frequency domains

This paper attempts to estimate the probability distribution of music signals. A number of music signals belonging to different genres of music have been analyzed. Four well known speech distributions viz. Gaussian, Generalized Gamma, Laplacian and Cauchy have been tested as hypotheses. The distribution estimation has been carried out in time and Discrete-Cosine-Transform (DCT) domains. It was observed that skewed Laplacian distribution describes the music samples most accurately with the peakedness of the distribution being correlated with the genre of music. Although Cauchy distribution along with Laplacian has been a good fit for most of the data, it is analytically shown in this work that Laplacian distribution is a better choice for modeling music signals.