Electrical - Electronic Engineering / Elektrik - Elektronik Mühendisliği
Permanent URI for this collectionhttps://hdl.handle.net/11147/11
Browse
4 results
Search Results
Conference Object Entropi Tabanlı Öznitelikler ile Müzik Enstrümanlarının Sınıflandırılması(IEEE, 2011) Özbek, Mehmet ErdalBy using the entropy values computed in temporal and spectral domain, the variations of musical signals can also be followed. In this study, the normalized entropy values computed in both domains are proposed to be used as features. These entropy based features are compared with similar features like temporal and spectral centroid, spectral flatness, and spectral spread. Then, their performances are investigated for classification of musical instruments. © 2011 IEEE.Article Skewed Alpha-Stable Distributions for Modeling and Classification of Musical Instruments(Türkiye Klinikleri Journal of Medical Sciences, 2012) Özbek, Mehmet Erdal; Çek, Mehmet Emre; Savacı, Ferit AcarMusic information retrieval and particularly musical instrument classification has become a very popular research area for the last few decades. Although in the literature many feature sets have been proposed to represent the musical instrument sounds, there is still need to find a superior feature set to achieve better classification performance. In this paper, we propose to use the parameters of skewed alpha-stable distribution of sub-band wavelet coefficients of musical sounds as features and show the effectiveness of this new feature set for musical instrument classification. We compare the classification performance with the features constructed from the parameters of generalized Gaussian density and some of the state-of-the-art features using support vector machine classifiers.Article Citation - WoS: 8Citation - Scopus: 8Wavelet Ridges for Musical Instrument Classification(Springer Verlag, 2012) Özbek, Mehmet Erdal; Özkurt, Nalan; Savacı, Ferit AcarThe time-varying frequency structure of musical signals have been analyzed using wavelets by either extracting the instantaneous frequency of signals or building features from the energies of sub-band coefficients. We propose to benefit from a combination of these two approaches and use the time-frequency domain energy localization curves, called as wavelet ridges, in order to build features for classification of musical instrument sounds. We evaluated the representative capability of our feature in different musical instrument classification problems using support vector machine classifiers. The comparison with the features based on parameterizing the wavelet sub-band energies confirmed the effectiveness of the proposed feature. © 2011 Springer Science+Business Media, LLC.Conference Object Citation - Scopus: 5Musical Note and Instrument Classification With Likelihood-Frequency Analysis and Support Vector Machines(Institute of Electrical and Electronics Engineers Inc., 2007) Özbek, Mehmet Erdal; Delpha, Claude; Duhamel, PierreIn this paper, we analyze the classification performance of a likelihood-frequency-time (LiFT) analysis designed for partial tracking and automatic transcription of music using support vector machines. The LiFT analysis is based on constant-Q filtering of signals with a filter-bank designed to filter 24 quarter-tone frequencies of an octave. Using the LiFT information, features are extracted from the isolated note samples and classification of instruments and notes is performed with linear, polynomial and radial basis function kernels. Correct classification ratios are obtained for 19 instrument and 36 notes.