Deep Learning Based EDM Subgenre Classification using Mel-Spectrogram and Tempogram Features
Published in
arXiv.org(2021)External Links:
TL
TL;DR
The state-of-art music auto-tagging model is extended to EDM subgenre classification, with the addition of two mid-level tempo-related feature representations, called the Fourier tempogram and autocorrelation tempogram, and two fusion strategies, early fusion and late fusion, are explored to aggregate the two types of tempograms.
Abstract
Along with the evolution of music technology, a large number of styles, or"subgenres,"of Electronic Dance Music(EDM) have emerged in recent years. While the classification task of distinguishing between EDM and non-EDM has been often studied in the context of music genre classification, little work has been done on the more challenging EDM subgenre classification. The state-of-art model is based on extremely randomized trees and could be improved by deep learning methods. In this paper, we extend the state-of-art music auto-tagging model"short-chunkCNN+Resnet"to EDM subgenre classification, with the addition of two mid-level tempo-related feature representations, called the Fourier tempogram and autocorrelation tempogram. And, we explore two fusion strategies, early fusion and late fusion, to aggregate the two types of tempograms. We evaluate the proposed models using a large dataset consisting of 75,000 songs for 30 different EDM subgenres, and show that the adoption of deep learning models and tempo features indeed leads to higher classification accuracy.
References27 items
1
Automatic DJ Transitions with Differentiable Audio Effects and Generative Adversarial Networks
2
Multimodal Metric Learning for Tag-Based Music Retrieval
3
Evaluation of CNN-based Automatic Music Tagging Models
4
Automatic subgenre classification in an electronic dance music taxonomy
5