Recognizing song mood and theme using convolutional recurrent neural networks

Mayerl, MaximilianMaximilianMayerlVötter, MichaelMichaelVötterHung, Hsiao TzuHsiao TzuHungChen, Bo YuBo YuChenYI-HSUAN YANGZangerle, EvaEvaZangerle2023-10-192023-10-192019-01-0116130073https://scholars.lib.ntu.edu.tw/handle/123456789/636277In this year's MediaEval task, Emotion and Theme Recognition in Music Using Jamendo, the goal is to assign emotion and theme tags to songs. In this paper, we describe our-Team TaiInn (Innsbruck)-approach for this task. We use a neural network model consisting of both convolutional and recurrent layers and utilize spectral, high-level as well as rhythm features. Our approach achieves a ROC-AUC score of 0.723 on the provided test set.Recognizing song mood and theme using convolutional recurrent neural networksconference paper2-s2.0-85091587047https://api.elsevier.com/content/abstract/scopus_id/85091587047