COMPLEX NMF: A NEW SPARSE REPRESENTATION FOR ACOUSTIC SIGNALS

This paper presents a new sparse representation for acous-tic signals which is based on a mixing model defined in the complex-spectrum domain (where additivity holds), and al-lows us to extract recurrent patterns of magnitude spectra that underlie observed complex spectra and the phase estimates of constituent signals. An efficient iterative algorithm is de-rived, which reduces to the multiplicative update algorithm for non-negative matrix factorization developed by Lee under a particular condition. Index Terms — Sparse signal representation, non-negative matrix factorization, sparse coding, data-driven approac