Sample-level Deep Convolutional Neural Networks for Music Auto-tagging Using Raw Waveforms

TL;DR

This paper introduces sample-level deep convolutional neural networks that learn from very small waveform segments, significantly improving music auto-tagging accuracy and providing insights into learned hierarchical features.

Contribution

It proposes a novel sample-level CNN architecture for music auto-tagging, demonstrating improved accuracy and interpretability over existing methods.

Findings

Achieved state-of-the-art results on Magnatagatune and Million Song Dataset

Visualized hierarchical features sensitive to log-scaled frequency

Sample-level filters enhance representation learning in music signals

Abstract

Recently, the end-to-end approach that learns hierarchical representations from raw data using deep convolutional neural networks has been successfully explored in the image, text and speech domains. This approach was applied to musical signals as well but has been not fully explored yet. To this end, we propose sample-level deep convolutional neural networks which learn representations from very small grains of waveforms (e.g. 2 or 3 samples) beyond typical frame-level input representations. Our experiments show how deep architectures with sample-level filters improve the accuracy in music auto-tagging and they provide results comparable to previous state-of-the-art performances for the Magnatagatune dataset and Million Song Dataset. In addition, we visualize filters learned in a sample-level DCNN in each layer to identify hierarchically learned features and show that they are sensitive to log-scaled frequency along layer, such as mel-frequency spectrogram that is widely used in music classification systems.