Majorization-Minimization Algorithm for Discriminative Non-Negative Matrix Factorization
Open Access
- 18 December 2020
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 8, 227399-227408
Abstract
This paper proposes a basis training algorithm for discriminative non-negative matrix factorization (NMF) with applications to single-channel audio source separation. With an NMF-based approach to supervised audio source separation, NMF is first applied to train the basis spectra of each source using training examples and then applied to the spectrogram of a mixture signal using the pretrained basis spectra at test time. The source signals can then be separated out using a Wiener filter. Here, a typical way to train the basis spectra is to minimize the dissimilarity measure between the observed spectrogram and the NMF model. However, obtaining the basis spectra in this way does not ensure that the separated signal will be optimal at test time due to the inconsistency between the objective functions for training and separation (Wiener filtering). To address this mismatch, a framework called discriminative NMF (DNMF) has recently been proposed. While this framework is noteworthy in that it uses a common objective function for training and separation, the objective function becomes more analytically complex than that of regular NMF. In the original DNMF work, a multiplicative update algorithm was proposed for the basis training; however, the convergence of the algorithm is not guaranteed and can be very slow. To overcome this weakness, this paper proposes a convergence-guaranteed algorithm for DNMF based on a majorization-minimization principle. Experimental results show that the proposed algorithm outperform the conventional DNMF algorithm as well as the regular NMF algorithm in terms of both the signal-to-distortion and signal-to-interference ratios.Keywords
Funding Information
- JSPS KAKENHI (18J20059)
- JST CREST (JPMJCR19A3)
- Secom Science and Technology Foundation
This publication has 15 references indexed in Scilit:
- The 2016 Signal Separation Evaluation Campaign2017
- Discriminative non-negative matrix factorization with majorization-minimization2017
- Deep clustering: Discriminative embeddings for segmentation and separation2016
- Non-negative Matrix Factorization and Its Variants for Audio Signal Processing2016
- Target Source Separation Based on Discriminative Nonnegative Matrix Factorization Incorporating Cross-Reconstruction ErrorIEICE Transactions on Information and Systems, 2015
- Discriminative non-negative matrix factorization for single-channel speech separation2014
- Deep learning for monaural speech separation2014
- Supervised non-euclidean sparse NMF via bilevel optimization with applications to speech enhancement2014
- Learning a Discriminative Dictionary for Single-Channel Speech SeparationIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2014
- Nonnegative Matrix Factorization with the Itakura-Saito Divergence: With Application to Music AnalysisNeural Computation, 2009