Idetifying Complex Unknown Audio

As the internet search evolves toward multimedia content based search, audio content identification and retrieval will likely become one of the key components of next generation internet search machines. In this paper we develop and evaluate a methodology for the identification (classification) of unknown highly complex audio pieces from within the same genre (classical music). Our methodology combines a novel audio content descriptor, the wavelet dispersion vector, with neural net assessment of the similarity between unknown query vectors and known (example set) vectors. We define the wavelet dispersion vector as the histogram of the rank orders obtained by the wavelet coefficients of a given wavelet scale among all the coefficients (of all scales at a given time instant). We demonstrate that the wavelet dispersion vector precisely characterizes the audio content while achieving good generalization. We examine the identification performance of a combination of 39 different wavelets and three different types of neural nets. We find that our wavelet dispersion vector calculated with a biorthogonal wavelet in conjunction with a probabilistic radial basis neural net trained by only three independent example sets correctly identifies approximately 78% of the unknown audio query pieces.