Music information retrieval (MIR) is concerned with the extraction and analysis of musical features from audio samples or scores (musical notations) from texts. These musical features supply the quantified and graphical information from the musical data, that is to say, the data corresponding to the pitch, loudness, note onset, note duration, pitch movements between notes, timbral characteristics depicting the quality of sound, and so on. MIR also involves music indexing and devising search strategies, such as content-based search. Obviously, the advancement of computer technology and the availability of digitized scores have made MIR an important field of interest for music researchers, music composers, and performers. MIR makes the wide reservoir of music open for these people, not to speak of the music listeners.

This survey paper on MIR covers some of the latest developments in the field, such as semantic auto-tagging and user-centric retrieval and recommendation approaches. Some of the topics covered in the survey are the motivation and history of MIR, music modalities and representations, music content description and indexing (feature extraction, similarity analysis, classification, and auto tagging), context-based music description and indexing, computational modeling, music discovery systems, and the challenging research problem of evaluation in MIR.

The target readership includes, but is not limited to, both music researchers and musicians. Although the authors have missed a few references, such as Ras and Wieczorkowska [1], the reader will find this paper informative.