This paper focuses on cover song identification among datasets potentially containing millions of songs. A compact representation of music contents plays an important role in large-scale analysis and retrieval. The proposed approach is based on high-level summarization of musical songs using chord profiles. Search is performed in two steps. In the first step, the Locality Sensitive Hashing (LHS) method is used to retrieve songs with similar chord profiles. On the resulting list of songs a second processing step is applied to progressively refine the ranking. Experiments conducted on both the Million Song Dataset (MSD) and a subset of the Second Hand Songs (SHS) dataset showed the effectiveness of the proposed solution, which provides state-of-the-art results.
Large-scale cover song identification using chord profiles
Khadkevich, Maksim;Omologo, Maurizio
2013-01-01
Abstract
This paper focuses on cover song identification among datasets potentially containing millions of songs. A compact representation of music contents plays an important role in large-scale analysis and retrieval. The proposed approach is based on high-level summarization of musical songs using chord profiles. Search is performed in two steps. In the first step, the Locality Sensitive Hashing (LHS) method is used to retrieve songs with similar chord profiles. On the resulting list of songs a second processing step is applied to progressively refine the ranking. Experiments conducted on both the Million Song Dataset (MSD) and a subset of the Second Hand Songs (SHS) dataset showed the effectiveness of the proposed solution, which provides state-of-the-art results.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
