Multi-scale Embedded CNN for Music Tagging (MsE-CNN)
arXiv:1906.06746
Abstract
Convolutional neural networks (CNN) recently gained notable attraction in a variety of machine learning tasks: including music classification and style tagging. In this work, we propose implementing intermediate connections to the CNN architecture to facilitate the transfer of multi-scale/level knowledge between different layers. Our novel model for music tagging shows significant improvement in comparison to the proposed approaches in the literature, due to its ability to carry low-level timbral features to the last layer.
arXiv commentProceedings of the 36th International Conference on Machine Learning (ICML)
References in corpus (5)
- Automatic tagging using deep convolutional neural networks
- Sample-level Deep Convolutional Neural Networks for Music Auto-tagging Using Raw Waveforms
- End-to-end learning for music audio tagging at scale
- Learning Multiscale Features Directly From Waveforms
- A Deep Bag-of-Features Model for Music Auto-Tagging