Melody Detection in Polyphonic Audio: From Pitch Extraction to Note Identification

Authors

Abstract

The current digital music boom demands more flexible and user- centered analysis, search and retrieval tools. This calls for mechanisms able to look inside audio signals and extract high-level content from them. To respond to these and other challenges, music information retrieval (MIR) has emerged as a broad multi- disciplinary research field. Among its topics, melody detection in polyphonic audio plays an important role, with applicability in areas such as automatic transcription, query-by-melody or plagiarism detection. In this book, different melody detection approaches are analyzed and a new system is proposed. Topics such as pitch detection, determination of musical notes from pitch sequences and identification of melodic notes are covered, as well as aspects of auditory physiology and perception. This analysis contributes to unify the current efforts in this area. It should be particularly useful to researchers working in MIR in general and melody detection in particular, as support to (under)graduate courses in music technology programs or to anyone enthusiastic to start messing around in this exciting and prolific research area.

Keywords

melody detection in polyphonic audio, music information retrieval, melody perception, musicology, pitch detection, conversion of pitch sequences into musical notes, pitch tracking and temporal segmentation, onset detection, identification of melodic notes

Subject

Music Information Retrieval

Book

Melody Detection in Polyphonic Audio: From Pitch Extraction to Note Identification, 978-3838319704, Lambert Academic Publishing, November 2009