THE STANDARDIZED VARIOGRAM AS A NOVEL TOOL FOR AUDIO SIMILARITY MEASURE

Most of methods for audio similarity evaluation are based on the Mel frequency cepstral coefficients, employed as main tool for the characterization of audio contents. Such approach needs some way of data compression aimed to optimize the information retrieval task and to reduce the computational costs derived from the usage of cluster analysis tools and probabilistic models. A novel approach is presented in this paper, based on the standardized variogram. This tool, inherited from Geostatistics, is applied to MFCCs matrices to reduce their size and compute compact representations of the audio contents (song signatures), aimed to evaluate audio similarity. The performance of the proposed approach is analyzed in comparison with other alternative methods and on the base of human responses.

[1]  Jonathan Foote,et al.  Content-based retrieval of music and audio , 1997, Other Conferences.

[2]  J. Stephen Downie,et al.  The Music Information Retrieval Evaluation eXchange (MIREX) , 2006 .

[3]  M. Stephens.  EDF Statistics for Goodness of Fit and Some Comparisons , 1974 .

[4]  John Haslett,et al.  On the sample variogram and the sample autocovariance for non-stationary time series , 1997 .

[5]  D. Krige.  A statistical approach to some basic mine valuation problems on the Witwatersrand, by D.G. Krige, published in the Journal, December 1951 : introduction by the author , 1951 .

[6]  Davit Khachatryan,et al.  Some results on the variogram in time series analysis , 2009, Qual. Reliab. Eng. Int..

[7]  Michael Edward Hohn,et al.  An Introduction to Applied Geostatistics: by Edward H. Isaaks and R. Mohan Srivastava, 1989, Oxford University Press, New York, 561 p., ISBN 0-19-505012-6, ISBN 0-19-505013-4 (paperback), $55.00 cloth, $35.00 paper (US) , 1991 .

[8]  Identifying short-range and long-range structural components of a compacted soil: an integrated geostatistical and spectral approach☆ , 2003 .

[9]  Jean Schoentgen,et al.  Dysphonic speech analysis using generalized variogram , 2005, Proceedings. (ICASSP '05). IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005..

[10]  Beth Logan,et al.  Mel Frequency Cepstral Coefficients for Music Modeling , 2000, ISMIR.

[11]  Hans Wackernagel,et al.  Multivariate Geostatistics: An Introduction with Applications , 1996 .

[12]  François Pachet,et al.  Music Similarity Measures: What's the use? , 2002, ISMIR.

[13]  Beth Logan,et al.  A Content-Based Music Similarity Function , 2001 .