Emotion and Theme Recognition in Music with Frequency-Aware RF-Regularized CNNs
Sound
2019-11-15 v1 Machine Learning
Multimedia
Audio and Speech Processing
Abstract
We present CP-JKU submission to MediaEval 2019; a Receptive Field-(RF)-regularized and Frequency-Aware CNN approach for tagging music with emotion/mood labels. We perform an investigation regarding the impact of the RF of the CNNs on their performance on this dataset. We observe that ResNets with smaller receptive fields -- originally adapted for acoustic scene classification -- also perform well in the emotion tagging task. We improve the performance of such architectures using techniques such as Frequency Awareness and Shake-Shake regularization, which were used in previous work on general acoustic recognition tasks.
Cite
@article{arxiv.1911.05833,
title = {Emotion and Theme Recognition in Music with Frequency-Aware RF-Regularized CNNs},
author = {Khaled Koutini and Shreyan Chowdhury and Verena Haunschmid and Hamid Eghbal-zadeh and Gerhard Widmer},
journal= {arXiv preprint arXiv:1911.05833},
year = {2019}
}
Comments
MediaEval`19, 27-29 October 2019, Sophia Antipolis, France