University of Limerick
Browse

Time domain note average energy based music onset detection

Download (58.33 kB)
conference contribution
posted on 2022-10-05, 10:24 authored by Ruolun Liu, Niall Griffith, JACQUELINE WALKERJACQUELINE WALKER, Peter Murphy
A novel time domain strategy is proposed for the detection of the onset of musical notes based on the changing energy level. By calculating the note average energy (NAE), the proposed method is insensitive to both the dynamic range of the energy levels in a piece and whether or notthe piece is monophonic or polyphonic. More importantly, the new strategy tackles the thorny ‘threshold’ problem that is always being avoided unsuccessfully. The detection performance of the new method is illustrated by its performance over a range of music pieces played on different instruments

History

Publication

Stockholm Music Acoustics Conference (SMAC 03); pp. 553-556

Note

peer-reviewed

Language

English

Department or School

  • Allied Health
  • Computer Science & Information Systems

Usage metrics

    University of Limerick

    Categories

    No categories selected

    Keywords

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC