Loading...
Please wait, while we are loading the content...
Similar Documents
TRANSFER LEARNING BY SUPERVISED PRE-TRAINING FOR AUDIO-BASED MUSIC CLASSIFICATION
| Content Provider | CiteSeerX |
|---|---|
| Author | Oord, Aäron Van Den Dieleman, Er Schrauwen, Benjamin |
| Abstract | Very few large-scale music research datasets are publicly available. There is an increasing need for such datasets, be-cause the shift from physical to digital distribution in the music industry has given the listener access to a large body of music, which needs to be cataloged efficiently and be easily browsable. Additionally, deep learning and feature learning techniques are becoming increasingly popular for music information retrieval applications, and they typically require large amounts of training data to work well. In this paper, we propose to exploit an available large-scale music dataset, the Million Song Dataset (MSD), for classifica-tion tasks on other datasets, by reusing models trained on the MSD for feature extraction. This transfer learning ap-proach, which we refer to as supervised pre-training, was previously shown to be very effective for computer vision problems. We show that features learned from MSD audio fragments in a supervised manner, using tag labels and user listening data, consistently outperform features learned in an unsupervised manner in this setting, provided that the learned feature extractor is of limited complexity. We eval-uate our approach on the GTZAN, 1517-Artists, Unique and Magnatagatune datasets. 1. |
| File Format | |
| Access Restriction | Open |
| Subject Keyword | Learned Feature Extractor Feature Extraction Limited Complexity Tag Label Classifica-tion Task User Listening Data Music Information Retrieval Application Available Large-scale Music Dataset Deep Learning Million Song Dataset Large-scale Music Research Datasets Outperform Feature Computer Vision Problem Feature Learning Technique Msd Audio Fragment Training Data Digital Distribution Magnatagatune Datasets Unsupervised Manner Supervised Manner Large Body Supervised Pre-training Listener Access Large Amount Music Industry |
| Content Type | Text |
