TY - GEN
T1 - A majorization-minimization algorithm with projected gradient updates for time-domain spectrogram factorization
AU - Kagami, Hideaki
AU - Kameoka, Hirokazu
AU - Yukawa, Masahiro
N1 - Funding Information:
This work was supported by the Support Center for Advanced Telecommunications Technology Research (SCAT) and JSPS Grants-in-Aid (15K06081, 15K13986, 15H02757, 26730100).
Publisher Copyright:
© 2017 IEEE.
PY - 2017/6/16
Y1 - 2017/6/16
N2 - We previously introduced a framework called time-domain spectrogram factorization (TSF), which realizes nonnegative matrix factorization (NMF)-like source separation in the time domain. This framework is particularly noteworthy in that, while maintaining the ability of NMF to obtain a parts-based representation of magnitude spectra, it allows us to (i) circumvent the commonly made assumption with the NMF approach that the magnitude spectra of source components are additive and (ii) take account of the interdependence of the phase/amplitude components at different time-frequency points. In particular, the second factor has been overlooked despite its potential importance. Our previous study revealed that the conventional TSF algorithm was relatively slow due to large matrix inversions, and the early stopping of the algorithm often resulted in poor separation accuracy. To overcome this problem, this paper presents an iterative TSF solver using projected gradient updates. Simulation results show that the proposed TSF approach yields higher source separation performance than NMF and the other variants including the original TSF.
AB - We previously introduced a framework called time-domain spectrogram factorization (TSF), which realizes nonnegative matrix factorization (NMF)-like source separation in the time domain. This framework is particularly noteworthy in that, while maintaining the ability of NMF to obtain a parts-based representation of magnitude spectra, it allows us to (i) circumvent the commonly made assumption with the NMF approach that the magnitude spectra of source components are additive and (ii) take account of the interdependence of the phase/amplitude components at different time-frequency points. In particular, the second factor has been overlooked despite its potential importance. Our previous study revealed that the conventional TSF algorithm was relatively slow due to large matrix inversions, and the early stopping of the algorithm often resulted in poor separation accuracy. To overcome this problem, this paper presents an iterative TSF solver using projected gradient updates. Simulation results show that the proposed TSF approach yields higher source separation performance than NMF and the other variants including the original TSF.
KW - Audio source separation
KW - non-negative matrix factorization(NMF)
KW - projected gradient method
UR - http://www.scopus.com/inward/record.url?scp=85023758677&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85023758677&partnerID=8YFLogxK
U2 - 10.1109/ICASSP.2017.7952218
DO - 10.1109/ICASSP.2017.7952218
M3 - Conference contribution
AN - SCOPUS:85023758677
T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
SP - 561
EP - 565
BT - 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017
Y2 - 5 March 2017 through 9 March 2017
ER -