Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
New Variants of Nonnegative Matrix Factorization with Application to Speech Coding and Speech Enhancement
KTH, School of Electrical Engineering and Computer Science (EECS).
2019 (English)Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

In this thesis, new variants of nonnegative matrix factorization (NMF) based ona convolutional data model, -divergence and sparsication are developed andanalyzed. These NMF variants are collectively referred to as -CNMF. Commonsparsication techniques such as L1-norm minimization and elastic net arediscussed and a new regularizer is proposed. It is shown that the new regularizer,unlike the above-mentioned sparsication techniques, has control overthe number of active bases in the NMF dictionary. Moreover, the -CNMF isextended to multichannel signals: it learns a common dictionary by exploitingthe correlation between channels through a multichannel coecient matrix. Asa result, an algorithm for source separation based on multichannel -CNMF isdeveloped. The algorithm is further tested in a multilayer setting, in which thefrequency-shifted coecient matrices serve as input to the next higher layer.Finally, three variants of the algorithm are evaluated in the context of speechenhancement, focusing on the problem of speech extraction from complex auditoryscenes. Figures obtained from the SiSEC 2016 data show that the proposedalgorithms perform comparably or better than the state of the art.

Abstract [sv]

Den här rapporten behandlar utveckling och analys av nya varianter av icke-negativ matrisfaktorisering (eng: nonnegative matrix factorization, NMF), som baseras på en datormodell med faltning, β-divergens och glesa matriser. Dessa varianter av NMF:er kallas allmänt för β-CNMF:er, där C:et står för “convolutional”. Vidare diskuteras vanliga tekniker för regularisering, såsom L1-normminimering och elastiska nät, och en ny formulering för regularisering föreslås. Det visar sig att denna nya formulering, till skillnad från ovan nämnda regulariseringstekniker, möjliggör kontroll av antalet aktiva basfunktioner i NMF:ens bibliotek. Utöver detta så utökas även β-CNMF:en till att behandla multikanalsignaler genom att tränas på en gemensam bibliotek som utnyttjar korskorrelationen mellan kanalerna. Detta möjliggör utveckling av en algoritm för källseparation av multikanalsignaler. Vidare så testas algoritmen i multipla led, där frekvensskiftade koefficientmatriser i ett led utgör indata till nästa led. Slutligen så bedöms tre olika varianter av algoritmen för talförbättring, med fokus på extrahering av tal ur komplexa ljudmiljöer. Mätningar från SiSEC 2016 visar att den föreslagna algoritmen presterar lika bra eller överträffar nu-varande befintliga algoritmer.

Place, publisher, year, edition, pages
2019. , p. 65
Series
TRITA-EECS-EX ; 2018:659
National Category
Engineering and Technology
Identifiers
URN: urn:nbn:se:kth:diva-253264OAI: oai:DiVA.org:kth-253264DiVA, id: diva2:1324307
External cooperation
Dolby Sweden
Educational program
Master of Science - Wireless Systems
Examiners
Available from: 2019-06-13 Created: 2019-06-13 Last updated: 2019-06-13Bibliographically approved

Open Access in DiVA

fulltext(8436 kB)41 downloads
File information
File name FULLTEXT01.pdfFile size 8436 kBChecksum SHA-512
75055d6f1be64dd16f194771fbf68d81a191c26548de314995e729a3aff0b0296eaac474c39ffb1ec5c3be1944c1e37b8e3078790f9c6ad55ff3336c49e335d6
Type fulltextMimetype application/pdf

By organisation
School of Electrical Engineering and Computer Science (EECS)
Engineering and Technology

Search outside of DiVA

GoogleGoogle Scholar
Total: 41 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 178 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf