• DocumentCode
    178007
  • Title

    Towards complex matrix decomposition of spectrograms based on the relative phase offsets of harmonic sounds

  • Author

    Kirchhoff, Holger ; Badeau, Roland ; Dixon, Sam

  • Author_Institution
    Centre for Digital Music, Queen Mary Univ. of London, London, UK
  • fYear
    2014
  • fDate
    4-9 May 2014
  • Firstpage
    1572
  • Lastpage
    1576
  • Abstract
    In this paper we study the relative phase offsets between partials in the sustained part of harmonic sounds and investigate their suitability for complex matrix decomposition of spectrograms. We formally introduce this property in a sinusoidal model and visualise the phase relations of a musical instrument. A model of complex matrix decomposition in the time-frequency domain is derived and equations for the estimation of the model parameters are provided in the mono-phonic case. We illustrate the model with the analysis of a mono-phonic saxophone signal. The results suggest that the phase offset is able to capture inherent time-invariant phase properties of harmonic sounds and outline its potential use for complex matrix decomposition.
  • Keywords
    acoustic signal processing; matrix decomposition; musical instruments; parameter estimation; time-frequency analysis; complex matrix decomposition; harmonic sounds; model parameter estimation; monophonic saxophone signal analysis; musical instrument; phase relations; relative phase offsets; sinusoidal model; spectrograms; time-frequency domain; time-invariant phase property; Bars; Harmonic analysis; Instruments; Mathematical model; Matrix decomposition; Spectrogram; Time-frequency analysis; complex matrix decomposition; harmonic signals; nonnegative matrix factorisation; relative phase offsets of partials;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Acoustics, Speech and Signal Processing (ICASSP), 2014 IEEE International Conference on
  • Conference_Location
    Florence
  • Type

    conf

  • DOI
    10.1109/ICASSP.2014.6853862
  • Filename
    6853862