Automatic genre classification of music content: a survey

Abstract

The creation of huge databases coming from both restoration of existing analog archives and new content is demanding more and more reliable and fast tools for content analysis and description, to be used for searches, content queries and interactive access. In that context, musical genres are crucial descriptors since they have been widely used for years to organize music catalogues, libraries and music stores. Despite their use, musical genres remain a poorly defined concept, which make of the automatic classification problem a non-trivial task. In this article, we review the state-of-theart in automatic genre classification and present new directions in automatic organization of music collections.

Audio Cover Song Identification and Similarity: Background, Approaches, Evaluation, and Beyond

“…The sense of match implies different degrees of specificity: it can be exact, retrieving music with specific content, or approximate, retrieving near neighbors in a musical space where proximity encodes different senses of musical similarity [10]. One prototypical use case is genre classification [70]. In this case, one generally tries to group songs according to a commercially or culturally established label, the genre, where certain characteristics might be more or less the same but many others might radically change (category-based song grouping).…”

Musical Genre Classification Using SVM and Audio Features

“…Musical genres are categories that have arisen through a complex interplay of cultures, artists and market forces to characterize similarities between musicians or compositions and organize music collections [9]. Nowadays, music genres are often used to categorize music on radio, television and especially internet.…”

Music Genre Classification: A Semi-supervised Approach

“…The state-of-the-art music genre classification systems can be classified into those based on supervised or unsupervised approach [4].…”

“…These features are called low level features [15] because they usually describe sound on small scale, such as slices of 10 to 60 seconds. Spectral features have been used to distinguish between speech and music [4] and to identify isolated sounds [16] and instruments [17]. Rauber et al [6] used psychoacoustic features of music to determine similarities between music files.…”