Abstract
K-Means clustering and SVM (support vector machine) are both very different methods of classification. The purpose of the work discussed in this paper is to detect the played musical instrument, separately using K-Means clustering and SVM for various levels of clustering and classification. The research was started by detecting the onset in the audio signal to get the instances where the instrument(s) are played and then segregating them depending on the played instrument. The Mel frequency Cepstral Coefficients (MFCCs) are then collected and eventually graded to detect the instrument used with the assistance of K-Means and SVM. Finally, the results obtained by individual SVM and K-Means clustering are constantly compared to obtain more accurate result. It is assumed at the end that the difference in results is due to the fundamental difference between how SVM and K-Means function and identify the instances.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Li P, Qian J, Wang T. Automatic instrument recognition in polyphonic music using convolutional neural networks. arXiv:1511.05520, 2015.
Takahashi Y, Kondo K. Comparison of two classification methods for musical instrument identification. In: IEEE 3rd Global Conference on Consumer Electronics (GCCE), 2014; pp. 67–68.
Goto M, Hayamizu S. A real-time music scene description system: detecting melody and bass lines in audio signals. In: IJCAI-99 workshop on computational auditory scene analysis, Stockholm, 1999; pp. 31–40.
Bosch JJ, Janer J, Fuhrmann F, Herrera P. A comparison of sound segregation techniques for predominant instrument recognition in musical audio signals. In: Proc. ISMIR (pp. 559–564), 2012.
Vaseghi SV, Rayner PJW. Detection and suppression of impulsive noise in speech communication systems. IEE Proc I- Commun Speech Vis. 1990;137(1):38–46.
Goto M, Muraoka Y. Beat tracking based on multiple-agent architecture̵a real-time beat tracking system for audio signals. In: Proc. 2nd Int. Conf. Multiagent Systems, 1996;pp. 103–110.
Kubera E, Wieczorkowska A, Skrzypiec M. Influence of feature setson precision, recall, and accuracy of identification of musical instruments in audio recordings. InISMIS, pp. 204–213. Springer, 2014.
Rafii Z, Pardo B. Music/voice separation using the similarity matrix. In: Proc. 13th Int. Soc. for Music Info. Retrieval Conf., Porto, 2012, pp. 583–588.
Zapat J, Gomez E. Using voice suppression algorithms to improve beat tracking in the presence of highly predominant vocals. In: 2013 IEEE international conference on acoustics, speech and signal processing.
Hainsworth S, Malcolm M. Onset detection in musical audio signals. In: Int. Computer Music Conf. (ICMC), Singapore, 2003;pp. 136–166.
Bhalke D, Rama Rao C, Bormane D. Automatic musical instrument classification using fractional fourier transform based-mfcc features and counter propagation neural network. J Intell Inf Syst. 2015;1–22.
Mel Frequency Cepstral Coefficients for Music Modeling- Beth Logan Cambridge Research Laboratory. http://musicweb.ucsd.edu/~sdubnov/CATbox/Reader/logan00mel.pdf Accessed 5 November 2019.
Molau S, Pitz M, Schluter R, Ney H. Computing Mel-frequency cepstral coefficients on the power spectrum. In: IEEE conference on acoustics, speecch, and signak processing, Salt Lake City, United States of America, 2001.
Fang Z, Guoliang Z, Zhanjiang S. Comparison of different implementations of MFCC. J Comput Sci Technol. p. 1. Accessed 5 November 2019.
Lohne M. The computational complexity of the fast Fourier transform. https://folk.uio.no/mathialo/texts/fftcomplexity.pdf. Accessed on 5 November 2019.
Gupta S, Jaafar J, Wan Ahmad F, Bansal A. Feature extraction using_MFCC. https://archive.is/20130414065947/http://asadl.org/jasa/resource/1/jasman/v8/i3/p185_s1. Accessed on 5 November 2019.
FitzGerald D., Jaiswal R.: ‘Improved Stereo Instrumental Track Recovery using Median Nearest-Neighbour Inpainting’, 24th IET Irish Signals and Systems Conference (ISSC 2013), 2013 page
Klapuri A. Musical instrument recognition in polyphonic audio using missing feature approach. IEEE Trans Audio Speech Lang Process. 2013;21(9):1805–17.
Castellengo M. Acoustical analysis of initial transients in flute-like instruments. Acta Acustica. 1999;85(3):387–400.
Herrera-Boyer P, Peeters G, Dubnov S. Automatic classification of musical instrument sounds. J New Music Res. 2003;32(1).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Pandey, A., Nair, T.R. & Thomas, S.B. Combination of K-Means Clustering and Support Vector Machine for Instrument Detection. SN COMPUT. SCI. 3, 121 (2022). https://doi.org/10.1007/s42979-021-01011-x
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s42979-021-01011-x