Solving Time-Domain Audio Inverse Problems Using Nonnegative Tensor Factorization
Nonnegative matrix factorization (NMF) and nonnegative tensor factorization (NTF) are important tools for modeling nonnegative data, which gained increasing popularity in various fields, a significant one of which is audio processing. However, there are still many problems in audio processing, for w...
Gespeichert in:
| Veröffentlicht in: | IEEE transactions on signal processing 2018-11, Vol.66 (21), p.5604-5617 |
|---|---|
| Hauptverfasser: | , , |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 5617 |
|---|---|
| container_issue | 21 |
| container_start_page | 5604 |
| container_title | IEEE transactions on signal processing |
| container_volume | 66 |
| creator | Bilen, Cagdas Ozerov, Alexey Perez, Patrick |
| description | Nonnegative matrix factorization (NMF) and nonnegative tensor factorization (NTF) are important tools for modeling nonnegative data, which gained increasing popularity in various fields, a significant one of which is audio processing. However, there are still many problems in audio processing, for which the NMF (or NTF) model has not been successfully utilized. In this paper, we propose a new algorithm based on the NMF (and NTF) in the short-time Fourier domain for solving a large class of audio inverse problems with missing or corrupted time-domain samples. The proposed approach overcomes the difficulty of employing a model in the frequency domain to recover time-domain samples with the help of probabilistic modeling. Its performance is demonstrated for the following applications: audio declipping and declicking (never solved with NMF/NTF modeling prior to this paper); joint audio declipping/declicking and source separation (never solved with NMF/NTF modeling or any other method prior to this paper); and compressive sampling recovery and compressive sampling-based informed source separation (an extremely low complexity encoding scheme that is possible with the proposed approach and has never been proposed prior to this paper). |
| doi_str_mv | 10.1109/TSP.2018.2869113 |
| format | Article |
| fullrecord | <record><control><sourceid>proquest_RIE</sourceid><recordid>TN_cdi_ieee_primary_8458165</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>8458165</ieee_id><sourcerecordid>2117177604</sourcerecordid><originalsourceid>FETCH-LOGICAL-c2823-16d7f02bc29244915745026e8ea04098552b74b33535e99e2109ebd239a2a9bd3</originalsourceid><addsrcrecordid>eNo9kM1LQkEUxR9RkFn7oM2DVi2ezZ2vN7MUyxSkDBXaDfP0aiP6xmZUqL--EcXVvVx-53DuybJ7IC0Aop_Ho2GLElAtqqQGYBdZAzSHgvBSXqadCFYIVX5dZzcxLgkBzrVsZJ8jv9q7epGP3RqLF7-2rs7bu5nzeb_eY4iYD4OvVriO-SQewHdf17iwW7fHfIx19CHv2unWB_eXjr6-za7mdhXx7jSb2aT7Ou70isHHW7_THhRTqigrQM7KOaHVlGqaooAouSBUokJLONFKCFqVvGJMMIFaI01PYjWjTFtqdTVjzezp6PttV2YT3NqGX-OtM732wBxuBKTUioo9JPbxyG6C_9lh3Jql34U6xTMUoISylIQnihypafAxBpyfbYGYQ8kmlWwOJZtTyUnycJQ4RDzjigsFUrB_odp17Q</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2117177604</pqid></control><display><type>article</type><title>Solving Time-Domain Audio Inverse Problems Using Nonnegative Tensor Factorization</title><source>IEEE Electronic Library (IEL)</source><creator>Bilen, Cagdas ; Ozerov, Alexey ; Perez, Patrick</creator><creatorcontrib>Bilen, Cagdas ; Ozerov, Alexey ; Perez, Patrick</creatorcontrib><description>Nonnegative matrix factorization (NMF) and nonnegative tensor factorization (NTF) are important tools for modeling nonnegative data, which gained increasing popularity in various fields, a significant one of which is audio processing. However, there are still many problems in audio processing, for which the NMF (or NTF) model has not been successfully utilized. In this paper, we propose a new algorithm based on the NMF (and NTF) in the short-time Fourier domain for solving a large class of audio inverse problems with missing or corrupted time-domain samples. The proposed approach overcomes the difficulty of employing a model in the frequency domain to recover time-domain samples with the help of probabilistic modeling. Its performance is demonstrated for the following applications: audio declipping and declicking (never solved with NMF/NTF modeling prior to this paper); joint audio declipping/declicking and source separation (never solved with NMF/NTF modeling or any other method prior to this paper); and compressive sampling recovery and compressive sampling-based informed source separation (an extremely low complexity encoding scheme that is possible with the proposed approach and has never been proposed prior to this paper).</description><identifier>ISSN: 1053-587X</identifier><identifier>EISSN: 1941-0476</identifier><identifier>DOI: 10.1109/TSP.2018.2869113</identifier><identifier>CODEN: ITPRED</identifier><language>eng</language><publisher>New York: IEEE</publisher><subject>Computer Science ; expectation-maximization algorithms ; Factorization ; Inverse problems ; iterative algorithms ; maximum a posteriori estimation ; maximum likelihood estimation ; Modelling ; Quantization (signal) ; Sampling ; Separation ; Signal and Image Processing ; Signal processing algorithms ; Signal reconstruction ; signal restoration ; Source separation ; Tensile stress ; Time domain analysis ; Time measurement</subject><ispartof>IEEE transactions on signal processing, 2018-11, Vol.66 (21), p.5604-5617</ispartof><rights>Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2018</rights><rights>Distributed under a Creative Commons Attribution 4.0 International License</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c2823-16d7f02bc29244915745026e8ea04098552b74b33535e99e2109ebd239a2a9bd3</citedby><cites>FETCH-LOGICAL-c2823-16d7f02bc29244915745026e8ea04098552b74b33535e99e2109ebd239a2a9bd3</cites><orcidid>0000-0002-2176-2720 ; 0000-0003-4834-5166</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/8458165$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>230,314,776,780,792,881,27901,27902,54733</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/8458165$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc><backlink>$$Uhttps://inria.hal.science/hal-01669825$$DView record in HAL$$Hfree_for_read</backlink></links><search><creatorcontrib>Bilen, Cagdas</creatorcontrib><creatorcontrib>Ozerov, Alexey</creatorcontrib><creatorcontrib>Perez, Patrick</creatorcontrib><title>Solving Time-Domain Audio Inverse Problems Using Nonnegative Tensor Factorization</title><title>IEEE transactions on signal processing</title><addtitle>TSP</addtitle><description>Nonnegative matrix factorization (NMF) and nonnegative tensor factorization (NTF) are important tools for modeling nonnegative data, which gained increasing popularity in various fields, a significant one of which is audio processing. However, there are still many problems in audio processing, for which the NMF (or NTF) model has not been successfully utilized. In this paper, we propose a new algorithm based on the NMF (and NTF) in the short-time Fourier domain for solving a large class of audio inverse problems with missing or corrupted time-domain samples. The proposed approach overcomes the difficulty of employing a model in the frequency domain to recover time-domain samples with the help of probabilistic modeling. Its performance is demonstrated for the following applications: audio declipping and declicking (never solved with NMF/NTF modeling prior to this paper); joint audio declipping/declicking and source separation (never solved with NMF/NTF modeling or any other method prior to this paper); and compressive sampling recovery and compressive sampling-based informed source separation (an extremely low complexity encoding scheme that is possible with the proposed approach and has never been proposed prior to this paper).</description><subject>Computer Science</subject><subject>expectation-maximization algorithms</subject><subject>Factorization</subject><subject>Inverse problems</subject><subject>iterative algorithms</subject><subject>maximum a posteriori estimation</subject><subject>maximum likelihood estimation</subject><subject>Modelling</subject><subject>Quantization (signal)</subject><subject>Sampling</subject><subject>Separation</subject><subject>Signal and Image Processing</subject><subject>Signal processing algorithms</subject><subject>Signal reconstruction</subject><subject>signal restoration</subject><subject>Source separation</subject><subject>Tensile stress</subject><subject>Time domain analysis</subject><subject>Time measurement</subject><issn>1053-587X</issn><issn>1941-0476</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2018</creationdate><recordtype>article</recordtype><sourceid>RIE</sourceid><recordid>eNo9kM1LQkEUxR9RkFn7oM2DVi2ezZ2vN7MUyxSkDBXaDfP0aiP6xmZUqL--EcXVvVx-53DuybJ7IC0Aop_Ho2GLElAtqqQGYBdZAzSHgvBSXqadCFYIVX5dZzcxLgkBzrVsZJ8jv9q7epGP3RqLF7-2rs7bu5nzeb_eY4iYD4OvVriO-SQewHdf17iwW7fHfIx19CHv2unWB_eXjr6-za7mdhXx7jSb2aT7Ou70isHHW7_THhRTqigrQM7KOaHVlGqaooAouSBUokJLONFKCFqVvGJMMIFaI01PYjWjTFtqdTVjzezp6PttV2YT3NqGX-OtM732wBxuBKTUioo9JPbxyG6C_9lh3Jql34U6xTMUoISylIQnihypafAxBpyfbYGYQ8kmlWwOJZtTyUnycJQ4RDzjigsFUrB_odp17Q</recordid><startdate>20181101</startdate><enddate>20181101</enddate><creator>Bilen, Cagdas</creator><creator>Ozerov, Alexey</creator><creator>Perez, Patrick</creator><general>IEEE</general><general>The Institute of Electrical and Electronics Engineers, Inc. (IEEE)</general><general>Institute of Electrical and Electronics Engineers</general><scope>97E</scope><scope>RIA</scope><scope>RIE</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7SC</scope><scope>7SP</scope><scope>8FD</scope><scope>JQ2</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope><scope>1XC</scope><scope>VOOES</scope><orcidid>https://orcid.org/0000-0002-2176-2720</orcidid><orcidid>https://orcid.org/0000-0003-4834-5166</orcidid></search><sort><creationdate>20181101</creationdate><title>Solving Time-Domain Audio Inverse Problems Using Nonnegative Tensor Factorization</title><author>Bilen, Cagdas ; Ozerov, Alexey ; Perez, Patrick</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c2823-16d7f02bc29244915745026e8ea04098552b74b33535e99e2109ebd239a2a9bd3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2018</creationdate><topic>Computer Science</topic><topic>expectation-maximization algorithms</topic><topic>Factorization</topic><topic>Inverse problems</topic><topic>iterative algorithms</topic><topic>maximum a posteriori estimation</topic><topic>maximum likelihood estimation</topic><topic>Modelling</topic><topic>Quantization (signal)</topic><topic>Sampling</topic><topic>Separation</topic><topic>Signal and Image Processing</topic><topic>Signal processing algorithms</topic><topic>Signal reconstruction</topic><topic>signal restoration</topic><topic>Source separation</topic><topic>Tensile stress</topic><topic>Time domain analysis</topic><topic>Time measurement</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Bilen, Cagdas</creatorcontrib><creatorcontrib>Ozerov, Alexey</creatorcontrib><creatorcontrib>Perez, Patrick</creatorcontrib><collection>IEEE All-Society Periodicals Package (ASPP) 2005-present</collection><collection>IEEE All-Society Periodicals Package (ASPP) 1998-Present</collection><collection>IEEE Electronic Library (IEL)</collection><collection>CrossRef</collection><collection>Computer and Information Systems Abstracts</collection><collection>Electronics & Communications Abstracts</collection><collection>Technology Research Database</collection><collection>ProQuest Computer Science Collection</collection><collection>Advanced Technologies Database with Aerospace</collection><collection>Computer and Information Systems Abstracts Academic</collection><collection>Computer and Information Systems Abstracts Professional</collection><collection>Hyper Article en Ligne (HAL)</collection><collection>Hyper Article en Ligne (HAL) (Open Access)</collection><jtitle>IEEE transactions on signal processing</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Bilen, Cagdas</au><au>Ozerov, Alexey</au><au>Perez, Patrick</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Solving Time-Domain Audio Inverse Problems Using Nonnegative Tensor Factorization</atitle><jtitle>IEEE transactions on signal processing</jtitle><stitle>TSP</stitle><date>2018-11-01</date><risdate>2018</risdate><volume>66</volume><issue>21</issue><spage>5604</spage><epage>5617</epage><pages>5604-5617</pages><issn>1053-587X</issn><eissn>1941-0476</eissn><coden>ITPRED</coden><abstract>Nonnegative matrix factorization (NMF) and nonnegative tensor factorization (NTF) are important tools for modeling nonnegative data, which gained increasing popularity in various fields, a significant one of which is audio processing. However, there are still many problems in audio processing, for which the NMF (or NTF) model has not been successfully utilized. In this paper, we propose a new algorithm based on the NMF (and NTF) in the short-time Fourier domain for solving a large class of audio inverse problems with missing or corrupted time-domain samples. The proposed approach overcomes the difficulty of employing a model in the frequency domain to recover time-domain samples with the help of probabilistic modeling. Its performance is demonstrated for the following applications: audio declipping and declicking (never solved with NMF/NTF modeling prior to this paper); joint audio declipping/declicking and source separation (never solved with NMF/NTF modeling or any other method prior to this paper); and compressive sampling recovery and compressive sampling-based informed source separation (an extremely low complexity encoding scheme that is possible with the proposed approach and has never been proposed prior to this paper).</abstract><cop>New York</cop><pub>IEEE</pub><doi>10.1109/TSP.2018.2869113</doi><tpages>14</tpages><orcidid>https://orcid.org/0000-0002-2176-2720</orcidid><orcidid>https://orcid.org/0000-0003-4834-5166</orcidid><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | ISSN: 1053-587X |
| ispartof | IEEE transactions on signal processing, 2018-11, Vol.66 (21), p.5604-5617 |
| issn | 1053-587X 1941-0476 |
| language | eng |
| recordid | cdi_ieee_primary_8458165 |
| source | IEEE Electronic Library (IEL) |
| subjects | Computer Science expectation-maximization algorithms Factorization Inverse problems iterative algorithms maximum a posteriori estimation maximum likelihood estimation Modelling Quantization (signal) Sampling Separation Signal and Image Processing Signal processing algorithms Signal reconstruction signal restoration Source separation Tensile stress Time domain analysis Time measurement |
| title | Solving Time-Domain Audio Inverse Problems Using Nonnegative Tensor Factorization |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-04T22%3A59%3A05IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_RIE&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Solving%20Time-Domain%20Audio%20Inverse%20Problems%20Using%20Nonnegative%20Tensor%20Factorization&rft.jtitle=IEEE%20transactions%20on%20signal%20processing&rft.au=Bilen,%20Cagdas&rft.date=2018-11-01&rft.volume=66&rft.issue=21&rft.spage=5604&rft.epage=5617&rft.pages=5604-5617&rft.issn=1053-587X&rft.eissn=1941-0476&rft.coden=ITPRED&rft_id=info:doi/10.1109/TSP.2018.2869113&rft_dat=%3Cproquest_RIE%3E2117177604%3C/proquest_RIE%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2117177604&rft_id=info:pmid/&rft_ieee_id=8458165&rfr_iscdi=true |