Discriminative Layered Nonnegative Matrix Factorization for Speech Separation
Files
No files are attached to this item.
Date
2016-01-01
Type
Journal Title
17TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION (INTERSPEECH 2016), VOLS 1-5: UNDERSTANDING SPEECH PROCESSING IN HUMANS AND MACHINES
Volume, issue and pages
pp. 560–564
ISSN
2308-457X
Language
English
Abstract
This paper proposes a discriminative layered nonnegative matrix factorization (DL-NMF) for monaural speech separation. The standard NMF conducts the parts-based representation using a single-layer of bases which was recently upgraded to the layered NMF (L-NMF) where a tree of bases was estimated for multi-level or multi-aspect decomposition of a complex mixed signal. In this study, we develop the DL-NMF by extending the generative bases in L-NMF to the discriminative bases which are estimated according to a discriminative criterion. The discriminative criterion is conducted by optimizing the recovery of the mixed spectra from the separated spectra and minimizing the reconstruction errors between separated spectra and original source spectra. The experiments on single-channel speech separation show the superiority of DL-NMF to NMF and L-NMF in terms of the SDR, SIR and SAR measures.
Keywords
dictionary learningdiscriminative learningnonnegative matrix factorizationspeech separationCollections
Conferences PaperCite this item
Hsu, C.-C., Chi, T.-S., & Chien, J.-T. (2016). Discriminative Layered Nonnegative Matrix Factorization for Speech Separation. 17TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION (INTERSPEECH 2016), VOLS 1-5: UNDERSTANDING SPEECH PROCESSING IN HUMANS AND MACHINES, 560–564. https://doi.org/10.21437/Interspeech.2016-415
Citation generated in APA 7th edition style; please check it before use.
Share
Permanent link: https://ir.lib.nycu.edu.tw/handle/11536/146771