TY - GEN
T1 - Optimal cost function and magnitude power for NMF-based speech separation and music interpolation
AU - King, Brian
AU - Fevotte, Cédric
AU - Smaragdis, Paris
PY - 2012
Y1 - 2012
N2 - There has been a significant amount of research in new algorithms and applications for nonnegative matrix factorization, but relatively little has been published on practical considerations for real-world applications, such as choosing optimal parameters for a particular application. In this paper, we will look at two applications, single-channel source separation of speech and interpolating missing music data. We will present the optimal parameters found for the experiments as well as discuss how parameters affect performance.
AB - There has been a significant amount of research in new algorithms and applications for nonnegative matrix factorization, but relatively little has been published on practical considerations for real-world applications, such as choosing optimal parameters for a particular application. In this paper, we will look at two applications, single-channel source separation of speech and interpolating missing music data. We will present the optimal parameters found for the experiments as well as discuss how parameters affect performance.
KW - Nonnegative matrix factorization
KW - source separation
KW - spectrogram interpolation
UR - http://www.scopus.com/inward/record.url?scp=84870706588&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84870706588&partnerID=8YFLogxK
U2 - 10.1109/MLSP.2012.6349726
DO - 10.1109/MLSP.2012.6349726
M3 - Conference contribution
AN - SCOPUS:84870706588
SN - 9781467310260
T3 - IEEE International Workshop on Machine Learning for Signal Processing, MLSP
BT - 2012 IEEE International Workshop on Machine Learning for Signal Processing - Proceedings of MLSP 2012
T2 - 2012 22nd IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2012
Y2 - 23 September 2012 through 26 September 2012
ER -