NMF with time-frequency activations to model non stationary audio events

Romain Hennequin, Roland Badeau, Bertrand DavidICASSP 2010 · OpenReview

Abstract

Real world sounds often exhibit non-stationary spectral characteristics such as those produced by a harpsichord or a guitar. The classical Non-negative Matrix Factorization (NMF) needs a number of atoms to accurately decompose the spectrogram of such sounds. An extension of NMF is proposed hereafter which includes time-frequency activations based on ARMA modeling. This leads to an efficient single-atom decomposition for a single audio event. The new algorithm is tested on real audio data and shows promising results.
BibTeX
@inproceedings{hennequin2010nmf,
  author = {Romain Hennequin and Roland Badeau and Bertrand David},
  title = {{NMF with time-frequency activations to model non stationary audio events}},
  booktitle = {IEEE International Conference on Acoustics, Speech and Signal Processing},
  year = {2010},
  url = {https://openreview.net/forum?id=3rEPaaa6BcK},
}

Loading related content...

Figure