Music genre recognition using convolutional recurrent neural network architecture

Abstract

The genre is an abstract feature, but still, it is considered to be one of the important characteristics of music. Genre recognition forms an essential component for a large number of commercial music applications. Most of the existing music genre recognition algorithms are based on manual feature extraction techniques. These extracted features are used to develop a classifier model to identify the genre. However, in many cases, it has been observed that a set of features giving excellent accuracy fails to explain the underlying typical characteristics of music genres. It has also been observed that some of the features provide a satisfactory level of performance on a particular dataset but fail to provide similar performance on other datasets. Hence, each dataset mostly requires manual selection of appropriate acoustic features to achieve an adequate level of performance on it. In this paper, we propose a genre recognition algorithm that uses almost no handcrafted features. The convolutional recurrent neural network‐based model proposed in this study is trained on melspectrogram extracted from 3‐s duration audio clips taken from GTZAN dataset. The proposed model provides an accuracy of 85.36% on 10‐class genre classification. The same model has been trained and tested on 10 genres of MagnaTagATune dataset having 18,476 clips of 29‐s duration. The model has yielded an accuracy of 86.06%. The experimental results suggest that the proposed architecture with melspectrogram as input feature is capable of providing consistent performances across the different datasets

A Globally Regularized Joint Neural Architecture for Music Classification

“…They suggested convolutional recurrent neural network (CRNN) models for the music auto-tagging and genre classification. The hybrid nature of neural networks has also been used in [3][5] for music recognition tasks. They used batch normalization on the input feature map, but fail to perform well in the situation where statistics change for various time steps, as mentioned in [12].…”

“…Each of the datasets has some common and distinct features with associated labels. The first dataset is GTZAN which consists of 1000 samples with 10 different genres, each having 100 songs as "blues, classical, country, disco, hip-hop, jazz, metal, pop, reggae, and rock" with a duration of 30 sec each as in [3] [5]. All files are mp3 and encoded with the sample rate of 22050 Hz with a size of 16 bit with the mono channel.…”

“…These systems are capable enough for automatic feature extraction, free from feature biases, and beneficial in comparison with the traditional methods. Neural Networks such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs) are among the influential methods for music classification [3][4] [5]. CNNs are favorable to record the spatial dependencies concerning feature domain [6] [7], whereas RNNs reasonably handle temporal dependencies in sequential data [8].…”

Multi-Level Local Feature Coding Fusion for Music Genre Recognition

“…For methods based on representation learning, a bidirectional recurrent neural network with serial attention and parallelized attention is proposed to focus on details of the target area [15]. A CNN-RNN-cascaded deep learning model that uses almost no handcrafted features is proposed in [14]. In [12], a CNN-based architecture with multi-level and multiscale features [11] is extended by transfer learning.…”

“…The scatter transform and the transfer learning have been widely used for image and audio classification [3]- [6] but rarely used for MGR in combination. Various methods for building music genre classifiers have been studied, including support vector machines (SVM) [7]- [9], Gaussian processes [10], convolutional neural network (CNN) [11] [12] [13], recurrent neural network (RNN) [14], and long short-term memory (LSTM) [15]. It is proven that CNNs learning representation features yield better performance in comparison to LSTMs and traditional methods which extract handcrafted features [16].…”