Representation Learning of Music Using Artist, Album, and Track Information
This addresses the issue of annotation cost for music researchers, but it is incremental as it adapts existing supervised methods to new metadata types.
The paper tackled the problem of costly semantic labeling for music representation learning by using naturally annotated factual metadata like artist, album, and track information, and found that combining these metadata improves overall performance.
Supervised music representation learning has been performed mainly using semantic labels such as music genres. However, annotating music with semantic labels requires time and cost. In this work, we investigate the use of factual metadata such as artist, album, and track information, which are naturally annotated to songs, for supervised music representation learning. The results show that each of the metadata has individual concept characteristics, and using them jointly improves overall performance.