Music classification by low-rank semantic mappings

Abstract

A challenging open question in music classification is which music representation (i.e., audio features) and which machine learning algorithm is appropriate for a specific music classification task. To address this challenge, given a number of audio feature vectors for each training music recording that capture the different aspects of music (i.e., timbre, harmony, etc.), the goal is to find a set of linear mappings from several feature spaces to the semantic space spanned by the class indicator vectors. These mappings should reveal the common latent variables, which characterize a given set of classes and simultaneously define a multi-class linear classifier that classifies the extracted latent common features. Such a set of mappings is obtained, building on the notion of the maximum margin matrix factorization, by minimizing a weighted sum of nuclear norms. Since the nuclear norm imposes rank constraints to the learnt mappings, the proposed method is referred to as low-rank semantic mappings (LRSMs). The performance of the LRSMs in music genre, mood, and multi-label classification is assessed by conducting extensive experiments on seven manually annotated benchmark datasets. The reported experimental results demonstrate the superiority of the LRSMs over the classifiers that are compared to. Furthermore, the best reported classification results are comparable with or slightly superior to those obtained by the state-of-the-art task-specific music classification methods.

Music Genre Classification Using Masked Conditional Neural Networks

“…Their work reports the accuracy achieved on the Ballroom dataset using the same features (cortical representations) and the classifier used for the Homburg, where they achieved 81.93% on the Ballroom dataset. The work in [24] achieved 62.4% on the Homburg dataset using auditory cortical representations, MFCC and Chroma as features. A neural network based attempt on the Homburg dataset in the work of Schluter et al [31] achieved 45.5% using mcRBM [29], a variant of the RBM, applied on a mel-spectrogram.…”

Automatic genre classification of Indian Tamil and western music using fractional MFCC

“…Unsupervised models have no previous training, but predict the class by locating a cluster or a pattern in a given set of unknown labels. Gaussian mixture model (GMM), support vector machines (SVMs), K-nearest neighbours (KNNs) and neural networks are commonly used supervised classifier models for genre classification, since it is a single-label multiclass problem (Panagakis and Kotropoulos 2013). NTF (Non-negative Tensor factorization) classifiers (Benetos and Kotropoulos 2010), and sparse representation-classifiers (Chen and Ramadge 2013) have also been used in genre classification.…”

“…Baniya et al (2014) derived a feature set that included higher order moments of skewness, kurtosis and covariance of features in addition to their mean and variances, resulting in improvement of classification accuracy to 85.15 %. Panagakis and Kotropoulos (2013) proposed a novel classification method for single (genre and mood) and multi label (music tagging) by finding common latent variables that speeds up the learning process. BanitalebiDehkordi and Banitalebi-Dehkordi (2014) performed genre classification using sparse FFT based features that included both short term and long term features.…”

“…Manual tagging is tedious (Panagakis and Kotropoulos 2013), subjective and & Betsy Rajesh betsyrajesh@gmail.com greatly dependent on cultural backgrounds (Benetos and Kotropoulos 2010). Different countries have different forms of music, differing in style, rhythm and instrumentation; the same genre can have different definitions in various parts of the world.…”

Genre Classification for Brazilian Music using Independent and Discriminant Features

“…In order to allow fair comparison of the proposed system with previous research in automatic music genre classification, were used here the results presented in [33] for different databases such as GTZAN [4], ISMIR 2004 [34], Homburg [35] and 1,517 Artists [36].…”