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

Author

Hennequin, Romain ; Badeau, Roland ; David, Bertrand

Author_Institution

Inst. TELECOM, TELECOM ParisTech, Paris, France

fYear

2010

fDate

14-19 March 2010

Firstpage

445

Lastpage

448

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.

Keywords

acoustic signal processing; audio signal processing; autoregressive moving average processes; singular value decomposition; spectral analysis; time-frequency analysis; ARMA modeling; autoregressive moving average processes; music information retrieval; nonnegative matrix factorization; nonstationary audio events; single-atom decomposition; spectrogram; time-frequency activations; Autoregressive processes; Cost function; Filters; Instruments; Machine learning algorithms; Matrix decomposition; Signal processing algorithms; Spectrogram; Telecommunications; Time frequency analysis; music information retrieval; non-negative matrix factorization; unsupervised machine learning;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics Speech and Signal Processing (ICASSP), 2010 IEEE International Conference on

Conference_Location

Dallas, TX

ISSN

1520-6149

Print_ISBN

978-1-4244-4295-9

Electronic_ISBN

1520-6149

Type

conf

DOI

10.1109/ICASSP.2010.5495733

Filename

5495733