CFP last date
22 April 2024
Reseach Article

Classification of Music Genre using Neural Networks with Cross-Entropy Optimization and Soft-Max Output

by Dharin Shah, Chirag Sachdev, Bhavik Shah
International Journal of Computer Applications
Foundation of Computer Science (FCS), NY, USA
Volume 119 - Number 12
Year of Publication: 2015
Authors: Dharin Shah, Chirag Sachdev, Bhavik Shah
10.5120/21123-4013

Dharin Shah, Chirag Sachdev, Bhavik Shah . Classification of Music Genre using Neural Networks with Cross-Entropy Optimization and Soft-Max Output. International Journal of Computer Applications. 119, 12 ( June 2015), 33-38. DOI=10.5120/21123-4013

@article{ 10.5120/21123-4013,
author = { Dharin Shah, Chirag Sachdev, Bhavik Shah },
title = { Classification of Music Genre using Neural Networks with Cross-Entropy Optimization and Soft-Max Output },
journal = { International Journal of Computer Applications },
issue_date = { June 2015 },
volume = { 119 },
number = { 12 },
month = { June },
year = { 2015 },
issn = { 0975-8887 },
pages = { 33-38 },
numpages = {9},
url = { https://ijcaonline.org/archives/volume119/number12/21123-4013/ },
doi = { 10.5120/21123-4013 },
publisher = {Foundation of Computer Science (FCS), NY, USA},
address = {New York, USA}
}
%0 Journal Article
%1 2024-02-06T23:03:53.381622+05:30
%A Dharin Shah
%A Chirag Sachdev
%A Bhavik Shah
%T Classification of Music Genre using Neural Networks with Cross-Entropy Optimization and Soft-Max Output
%J International Journal of Computer Applications
%@ 0975-8887
%V 119
%N 12
%P 33-38
%D 2015
%I Foundation of Computer Science (FCS), NY, USA
Abstract

In this paper, an abstract model to predict the genre of a music audio file is proposed (specifically a wave file). The output of the model is the probability distribution along the considered genres. A machine learning approach is employed. The adaptive learning process is modeled by neural networks with back-propagation as its learning algorithm and cross entropy as its optimization function. The emphasis is on feature extractors since the learning paradigm is well known to other applications. Simple Analysis on the Features were performed for appropriate selection.

References
  1. O. Lartillot and P. Toiviainen, "A Matlab Toolbox for Musical Feature Extraction From Audio," International Conference on Digital Audio Effects, 2007.
  2. M. A. Nielsen, Neural Networks and Deep Learning, Determination Press, 2015.
  3. C. Mckay, "Using Neural Networks for Music Genre Classification," Faculty of Music.
  4. M. K. Shan and F. F. Kuo, "Music Style mining and classification by melody," IEICE Transactions on Information and Systems, Vols. E86- D(3). 655-659. .
  5. D. Cereghetti, O. Lartillot, K. Eliard, W. J. Trost, M. A. Rappaz and D. Grandjean, "Estimating tempo and metrical features by tracking the whole metrical hierarchy," 3rd International Conference on Music and Emotion, 2013.
  6. O. Lartillot, "Computational analysis of maqam music: From audio transcription to musicological analysis, everything is tightly intertwined," in Acoustics 2012 Hong Kong Conference, Hong Kong.
Index Terms

Computer Science
Information Sciences

Keywords

Auto-Encoder Classification Cross-Entropy Machine Learning Music Genre Neural Networks Soft-max Output.