A light-weight full-band speech enhancement model
Deep neural network based full-band speech enhancement systems face challenges of high demand of computational resources and imbalanced frequency distribution. In this paper, a light-weight full-band model is proposed with two dedicated strategies, i.e., a learnable spectral compression mapping for...
Gespeichert in:
| Veröffentlicht in: | arXiv.org 2022-07 |
|---|---|
| Hauptverfasser: | , , , |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | |
|---|---|
| container_issue | |
| container_start_page | |
| container_title | arXiv.org |
| container_volume | |
| creator | Hu, Qinwen Hou, Zhongshu Le, Xiaohuai Lu, Jing |
| description | Deep neural network based full-band speech enhancement systems face challenges of high demand of computational resources and imbalanced frequency distribution. In this paper, a light-weight full-band model is proposed with two dedicated strategies, i.e., a learnable spectral compression mapping for more effective high-band spectral information compression, and the utilization of the multi-head attention mechanism for more effective modeling of the global spectral pattern. Experiments validate the efficacy of the proposed strategies and show that the proposed model achieves competitive performance with only 0.89M parameters. |
| format | Article |
| fullrecord | <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2682587565</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2682587565</sourcerecordid><originalsourceid>FETCH-proquest_journals_26825875653</originalsourceid><addsrcrecordid>eNpjYuA0MjY21LUwMTLiYOAtLs4yMDAwMjM3MjU15mQwdFTIyUzPKNEtTwVRCmmlOTm6SYl5KQrFBampyRkKqXkZiXnJqbmpeSUKufkpqTk8DKxpiTnFqbxQmptB2c01xNlDt6Aov7A0tbgkPiu_tCgPKBVvZGZhZGphbmpmakycKgCCwDNU</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2682587565</pqid></control><display><type>article</type><title>A light-weight full-band speech enhancement model</title><source>Freely accessible journals*</source><creator>Hu, Qinwen ; Hou, Zhongshu ; Le, Xiaohuai ; Lu, Jing</creator><creatorcontrib>Hu, Qinwen ; Hou, Zhongshu ; Le, Xiaohuai ; Lu, Jing</creatorcontrib><description>Deep neural network based full-band speech enhancement systems face challenges of high demand of computational resources and imbalanced frequency distribution. In this paper, a light-weight full-band model is proposed with two dedicated strategies, i.e., a learnable spectral compression mapping for more effective high-band spectral information compression, and the utilization of the multi-head attention mechanism for more effective modeling of the global spectral pattern. Experiments validate the efficacy of the proposed strategies and show that the proposed model achieves competitive performance with only 0.89M parameters.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Artificial neural networks ; Business competition ; Frequency distribution ; Speech processing ; Weight reduction</subject><ispartof>arXiv.org, 2022-07</ispartof><rights>2022. This work is published under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>777,781</link.rule.ids></links><search><creatorcontrib>Hu, Qinwen</creatorcontrib><creatorcontrib>Hou, Zhongshu</creatorcontrib><creatorcontrib>Le, Xiaohuai</creatorcontrib><creatorcontrib>Lu, Jing</creatorcontrib><title>A light-weight full-band speech enhancement model</title><title>arXiv.org</title><description>Deep neural network based full-band speech enhancement systems face challenges of high demand of computational resources and imbalanced frequency distribution. In this paper, a light-weight full-band model is proposed with two dedicated strategies, i.e., a learnable spectral compression mapping for more effective high-band spectral information compression, and the utilization of the multi-head attention mechanism for more effective modeling of the global spectral pattern. Experiments validate the efficacy of the proposed strategies and show that the proposed model achieves competitive performance with only 0.89M parameters.</description><subject>Artificial neural networks</subject><subject>Business competition</subject><subject>Frequency distribution</subject><subject>Speech processing</subject><subject>Weight reduction</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><sourceid>ABUWG</sourceid><sourceid>AFKRA</sourceid><sourceid>AZQEC</sourceid><sourceid>BENPR</sourceid><sourceid>CCPQU</sourceid><sourceid>DWQXO</sourceid><recordid>eNpjYuA0MjY21LUwMTLiYOAtLs4yMDAwMjM3MjU15mQwdFTIyUzPKNEtTwVRCmmlOTm6SYl5KQrFBampyRkKqXkZiXnJqbmpeSUKufkpqTk8DKxpiTnFqbxQmptB2c01xNlDt6Aov7A0tbgkPiu_tCgPKBVvZGZhZGphbmpmakycKgCCwDNU</recordid><startdate>20220703</startdate><enddate>20220703</enddate><creator>Hu, Qinwen</creator><creator>Hou, Zhongshu</creator><creator>Le, Xiaohuai</creator><creator>Lu, Jing</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope></search><sort><creationdate>20220703</creationdate><title>A light-weight full-band speech enhancement model</title><author>Hu, Qinwen ; Hou, Zhongshu ; Le, Xiaohuai ; Lu, Jing</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_26825875653</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2022</creationdate><topic>Artificial neural networks</topic><topic>Business competition</topic><topic>Frequency distribution</topic><topic>Speech processing</topic><topic>Weight reduction</topic><toplevel>online_resources</toplevel><creatorcontrib>Hu, Qinwen</creatorcontrib><creatorcontrib>Hou, Zhongshu</creatorcontrib><creatorcontrib>Le, Xiaohuai</creatorcontrib><creatorcontrib>Lu, Jing</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science & Engineering Collection</collection><collection>ProQuest Central (Alumni)</collection><collection>ProQuest Central</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Publicly Available Content Database (Proquest) (PQ_SDU_P3)</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering collection</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Hu, Qinwen</au><au>Hou, Zhongshu</au><au>Le, Xiaohuai</au><au>Lu, Jing</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>A light-weight full-band speech enhancement model</atitle><jtitle>arXiv.org</jtitle><date>2022-07-03</date><risdate>2022</risdate><eissn>2331-8422</eissn><abstract>Deep neural network based full-band speech enhancement systems face challenges of high demand of computational resources and imbalanced frequency distribution. In this paper, a light-weight full-band model is proposed with two dedicated strategies, i.e., a learnable spectral compression mapping for more effective high-band spectral information compression, and the utilization of the multi-head attention mechanism for more effective modeling of the global spectral pattern. Experiments validate the efficacy of the proposed strategies and show that the proposed model achieves competitive performance with only 0.89M parameters.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext |
| identifier | EISSN: 2331-8422 |
| ispartof | arXiv.org, 2022-07 |
| issn | 2331-8422 |
| language | eng |
| recordid | cdi_proquest_journals_2682587565 |
| source | Freely accessible journals* |
| subjects | Artificial neural networks Business competition Frequency distribution Speech processing Weight reduction |
| title | A light-weight full-band speech enhancement model |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-17T15%3A23%3A36IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=document&rft.atitle=A%20light-weight%20full-band%20speech%20enhancement%20model&rft.jtitle=arXiv.org&rft.au=Hu,%20Qinwen&rft.date=2022-07-03&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2682587565%3C/proquest%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2682587565&rft_id=info:pmid/&rfr_iscdi=true |