On the use of memory for detecting musical notes in polyphonic piano music

Author

Costantini, Giovanni ; Todisco, Massimiliano ; Perfetti, Renzo

Author_Institution

Dept. of Electron. Eng., Univ. of Rome, Rome, Italy

fYear

2009

Firstpage

806

Lastpage

809

Abstract

Music transcription consists in transforming the musical content of audio data into a symbolic representation. The objective of this study is to investigate a transcription system for polyphonic piano. The proposed method focuses on temporal musical structures, note events and their main characteristics: the attack instant and the pitch. Onset detection exploits a time-frequency representation of the audio signal. Note classification is based on constant Q transform (CQT) and support vector machines (SVMs). Finally, to validate our method, we present a collection of experiments using a wide number of musical pieces of heterogeneous style.

Keywords

audio signal processing; music; signal classification; support vector machines; time-frequency analysis; transforms; audio data; constant Q transform; music transcription; musical note detection; note classification; note events; onset detection; polyphonic piano music; support vector machines; symbolic representation; temporal musical structures; time-frequency representation; Acoustic signal detection; Acoustical engineering; Autocorrelation; Data engineering; Detection algorithms; Discrete Fourier transforms; Music; Support vector machine classification; Support vector machines; Time frequency analysis;

fLanguage

English

Publisher

ieee

Conference_Titel

Circuit Theory and Design, 2009. ECCTD 2009. European Conference on

Conference_Location

Antalya

Print_ISBN

978-1-4244-3896-9

Electronic_ISBN

978-1-4244-3896-9

Type

conf

DOI

10.1109/ECCTD.2009.5275106

Filename

5275106