{"id":"https://openalex.org/W3163105696","doi":"https://doi.org/10.1109/icassp39728.2021.9413536","title":"Compressing Deep Neural Networks for Efficient Speech Enhancement","display_name":"Compressing Deep Neural Networks for Efficient Speech Enhancement","publication_year":2021,"publication_date":"2021-05-13","ids":{"openalex":"https://openalex.org/W3163105696","doi":"https://doi.org/10.1109/icassp39728.2021.9413536","mag":"3163105696"},"language":"en","primary_location":{"id":"doi:10.1109/icassp39728.2021.9413536","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp39728.2021.9413536","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5058209019","display_name":"Ke Tan","orcid":"https://orcid.org/0000-0001-5073-8060"},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ke Tan","raw_affiliation_strings":["The Ohio State University, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Ohio State University, USA","institution_ids":["https://openalex.org/I52357470"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5051837453","display_name":"DeLiang Wang","orcid":"https://orcid.org/0000-0001-8195-6319"},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"DeLiang Wang","raw_affiliation_strings":["Center for Cognitive and Brain Sciences, The Ohio State University, USA","The Ohio State University, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Center for Cognitive and Brain Sciences, The Ohio State University, USA","institution_ids":["https://openalex.org/I52357470"]},{"raw_affiliation_string":"The Ohio State University, USA","institution_ids":["https://openalex.org/I52357470"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I52357470"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":14,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"8358","last_page":"8362"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10860","display_name":"Speech and Audio Processing","score":1.0,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10860","display_name":"Speech and Audio Processing","score":1.0,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10326","display_name":"Indoor and Outdoor Localization Technologies","score":0.994700014591217,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11233","display_name":"Advanced Adaptive Filtering Techniques","score":0.9937000274658203,"subfield":{"id":"https://openalex.org/subfields/2206","display_name":"Computational Mechanics"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/pesq","display_name":"PESQ","score":0.8692125082015991},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7955845594406128},{"id":"https://openalex.org/keywords/deep-neural-networks","display_name":"Deep neural networks","score":0.6779074668884277},{"id":"https://openalex.org/keywords/speech-enhancement","display_name":"Speech enhancement","score":0.6758118271827698},{"id":"https://openalex.org/keywords/regularization","display_name":"Regularization (linguistics)","score":0.5675211548805237},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5441568493843079},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.5253357887268066},{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.5132608413696289},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.4441901445388794},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.430536150932312},{"id":"https://openalex.org/keywords/speech-coding","display_name":"Speech coding","score":0.4256500005722046},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.39000362157821655},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.1830962896347046}],"concepts":[{"id":"https://openalex.org/C103734657","wikidata":"https://www.wikidata.org/wiki/Q2739975","display_name":"PESQ","level":4,"score":0.8692125082015991},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7955845594406128},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.6779074668884277},{"id":"https://openalex.org/C2776182073","wikidata":"https://www.wikidata.org/wiki/Q7575395","display_name":"Speech enhancement","level":3,"score":0.6758118271827698},{"id":"https://openalex.org/C2776135515","wikidata":"https://www.wikidata.org/wiki/Q17143721","display_name":"Regularization (linguistics)","level":2,"score":0.5675211548805237},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5441568493843079},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.5253357887268066},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.5132608413696289},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.4441901445388794},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.430536150932312},{"id":"https://openalex.org/C13895895","wikidata":"https://www.wikidata.org/wiki/Q3270773","display_name":"Speech coding","level":2,"score":0.4256500005722046},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.39000362157821655},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.1830962896347046},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.0},{"id":"https://openalex.org/C163294075","wikidata":"https://www.wikidata.org/wiki/Q581861","display_name":"Noise reduction","level":2,"score":0.0},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp39728.2021.9413536","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp39728.2021.9413536","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":47,"referenced_works":["https://openalex.org/W160800111","https://openalex.org/W1552314771","https://openalex.org/W1690739335","https://openalex.org/W1724438581","https://openalex.org/W1821462560","https://openalex.org/W1974387177","https://openalex.org/W1987371344","https://openalex.org/W1993482030","https://openalex.org/W2114766824","https://openalex.org/W2119144962","https://openalex.org/W2141998673","https://openalex.org/W2145085734","https://openalex.org/W2168379380","https://openalex.org/W2279098554","https://openalex.org/W2460144244","https://openalex.org/W2508418541","https://openalex.org/W2612445135","https://openalex.org/W2736953746","https://openalex.org/W2809624076","https://openalex.org/W2906544045","https://openalex.org/W2937484199","https://openalex.org/W2948116732","https://openalex.org/W2962866211","https://openalex.org/W2963122961","https://openalex.org/W2963125010","https://openalex.org/W2963864497","https://openalex.org/W2964118293","https://openalex.org/W2964299589","https://openalex.org/W2972346666","https://openalex.org/W2991361823","https://openalex.org/W2998784869","https://openalex.org/W3012561096","https://openalex.org/W3103034221","https://openalex.org/W4289099280","https://openalex.org/W4297775537","https://openalex.org/W6637551013","https://openalex.org/W6637709462","https://openalex.org/W6638523607","https://openalex.org/W6677103964","https://openalex.org/W6677580257","https://openalex.org/W6695314431","https://openalex.org/W6722866051","https://openalex.org/W6737664043","https://openalex.org/W6740474667","https://openalex.org/W6753069482","https://openalex.org/W6757218482","https://openalex.org/W6761206160"],"related_works":["https://openalex.org/W2058482658","https://openalex.org/W3016109656","https://openalex.org/W3135613579","https://openalex.org/W1973895194","https://openalex.org/W4388016426","https://openalex.org/W4386746628","https://openalex.org/W2546593254","https://openalex.org/W1980687383","https://openalex.org/W2166831097","https://openalex.org/W3209446892"],"abstract_inverted_index":{"The":[0],"use":[1],"of":[2,12,34,89,108,152],"deep":[3],"neural":[4],"networks":[5],"(DNNs)":[6],"has":[7,139],"dramatically":[8],"improved":[9],"the":[10,16,106,144,150,154],"performance":[11],"speech":[13,81],"enhancement":[14,29,115],"in":[15,58],"past":[17],"decade.":[18],"However,":[19],"a":[20,72,124,135,140],"large":[21,125],"DNN":[22,78,126,137],"is":[23,36,45,84],"typically":[24],"required":[25],"to":[26,47,66,76,143],"achieve":[27],"strong":[28],"performance,":[30],"and":[31,40,95,122,130],"this":[32,68],"kind":[33],"model":[35,73,156],"both":[37],"computationally":[38],"intensive":[39],"memory":[41],"consuming.":[42],"Hence":[43],"it":[44],"difficult":[46],"deploy":[48],"such":[49],"DNNs":[50,110],"on":[51,86],"devices":[52],"with":[53,60],"limited":[54],"hardware":[55],"resources":[56],"or":[57],"applications":[59],"strict":[61],"latency":[62],"requirements.":[63],"In":[64],"order":[65],"address":[67],"problem,":[69],"we":[70,118],"propose":[71],"compression":[74,157],"pipeline":[75],"reduce":[77],"size":[79,142],"for":[80],"enhancement,":[82],"which":[83],"based":[85],"three":[87],"kinds":[88],"techniques:":[90],"sparse":[91],"regularization,":[92],"iterative":[93],"pruning":[94],"clustering-based":[96],"quantization.":[97],"Evaluation":[98],"results":[99],"show":[100],"that":[101,120,138],"our":[102],"approach":[103],"substantially":[104],"reduces":[105],"sizes":[107],"different":[109],"without":[111],"significantly":[112],"affecting":[113],"their":[114],"performance.":[116],"Moreover,":[117],"find":[119],"training":[121,134],"compressing":[123],"yields":[127],"higher":[128],"STOI":[129],"PESQ":[131],"than":[132],"directly":[133],"small":[136],"comparable":[141],"compressed":[145],"DNN.":[146],"This":[147],"further":[148],"suggests":[149],"benefits":[151],"using":[153],"proposed":[155],"approach.":[158]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":3},{"year":2023,"cited_by_count":5},{"year":2022,"cited_by_count":2},{"year":2021,"cited_by_count":3}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}