Novelty Detection Based on Spectral Similarity of Songs

We are introducing novelty detection, i.e. the automatic identification of new or unknown data not covered by the training data, to the field of music information retrieval. Two methods for novelty detection - one based solely on the similarity information and one also utilizing genre label information - are evaluated within the context of genre classification based on spectral similarity. Both are shown to perform equally well.

• Type: Conference
• Year: 2005
• Where: ISMIR
• Authors: Arthur Flexer, Elias Pampalk, Gerhard Widmer
• Keywords: Automatic Identification, Genre Label Information, Information Retrieval, ISMIR 2005, Music Information Retrieval