MusicRainbow is a simple interface for discovering artists where colors encode different types of music. MusicRainbow is based on a new audio-based approach to compute artist similarity. This approach scores 15 percentage points higher in a genre classification task than the similarity computed on track level. Using a traveling salesman algorithm, similar artists are mapped near each other on a circular rainbow. Words describe different regions of the rainbow. These words are automatically extracted from web pages mentioning the artists. The authors used different vocabularies for different hierarchical levels and heuristics to select the most descriptive labels.The demonstration available online is based on a collection containing 558 artists. In this same demo we can see how the user controls the MusicRainbow interface with a knob which can be turned (to select an artist) and pushed (to listen to music from the selected artist).
