{"id":"https://openalex.org/W4317038458","doi":"https://doi.org/10.1109/tvlsi.2023.3235760","title":"A 1.6-mW Sparse Deep Learning Accelerator for Speech Separation","display_name":"A 1.6-mW Sparse Deep Learning Accelerator for Speech Separation","publication_year":2023,"publication_date":"2023-01-17","ids":{"openalex":"https://openalex.org/W4317038458","doi":"https://doi.org/10.1109/tvlsi.2023.3235760"},"language":"en","primary_location":{"id":"doi:10.1109/tvlsi.2023.3235760","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/tvlsi.2023.3235760","pdf_url":null,"source":{"id":"https://openalex.org/S37538908","display_name":"IEEE Transactions on Very Large Scale Integration (VLSI) Systems","issn_l":"1063-8210","issn":["1063-8210","1557-9999"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Very Large Scale Integration (VLSI) Systems","raw_type":"journal-article"},"type":"article","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2312.09580","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5039735116","display_name":"Chih-Chyau Yang","orcid":"https://orcid.org/0000-0001-6508-8160"},"institutions":[{"id":"https://openalex.org/I148366613","display_name":"National Yang Ming Chiao Tung University","ror":"https://ror.org/00se2k293","country_code":"TW","type":"education","lineage":["https://openalex.org/I148366613"]},{"id":"https://openalex.org/I4210120917","display_name":"Taiwan Semiconductor Manufacturing Company (Taiwan)","ror":"https://ror.org/02wx79d08","country_code":"TW","type":"company","lineage":["https://openalex.org/I4210120917"]}],"countries":["TW"],"is_corresponding":true,"raw_author_name":"Chih-Chyau Yang","raw_affiliation_strings":["Institute of Electronics, National Yang Ming Chiao Tung University, Hsinchu, Taiwan","Taiwan Semiconductor Research Institute, Hsinchu, Taiwan"],"affiliations":[{"raw_affiliation_string":"Institute of Electronics, National Yang Ming Chiao Tung University, Hsinchu, Taiwan","institution_ids":["https://openalex.org/I148366613"]},{"raw_affiliation_string":"Taiwan Semiconductor Research Institute, Hsinchu, Taiwan","institution_ids":["https://openalex.org/I4210120917"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5067793643","display_name":"Tian\u2010Sheuan Chang","orcid":"https://orcid.org/0000-0002-0561-8745"},"institutions":[{"id":"https://openalex.org/I148366613","display_name":"National Yang Ming Chiao Tung University","ror":"https://ror.org/00se2k293","country_code":"TW","type":"education","lineage":["https://openalex.org/I148366613"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Tian-Sheuan Chang","raw_affiliation_strings":["Institute of Electronics, National Yang Ming Chiao Tung University, Hsinchu, Taiwan"],"affiliations":[{"raw_affiliation_string":"Institute of Electronics, National Yang Ming Chiao Tung University, Hsinchu, Taiwan","institution_ids":["https://openalex.org/I148366613"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5039735116"],"corresponding_institution_ids":["https://openalex.org/I148366613","https://openalex.org/I4210120917"],"apc_list":null,"apc_paid":null,"fwci":0.6003,"has_fulltext":true,"cited_by_count":3,"citation_normalized_percentile":{"value":0.6174904,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":96,"max":97},"biblio":{"volume":"31","issue":"3","first_page":"310","last_page":"319"},"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.9991000294685364,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"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/T11447","display_name":"Blind Source Separation Techniques","score":0.9991000294685364,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7410464286804199},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.5720720291137695},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.5447223782539368},{"id":"https://openalex.org/keywords/floating-point","display_name":"Floating point","score":0.5233449339866638},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.4933909475803375},{"id":"https://openalex.org/keywords/kernel","display_name":"Kernel (algebra)","score":0.4483276903629303},{"id":"https://openalex.org/keywords/power","display_name":"Power (physics)","score":0.4249325394630432},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.39714333415031433},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.3616896867752075},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.3352113962173462},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.32561981678009033},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.26652562618255615},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.09187355637550354}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7410464286804199},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.5720720291137695},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.5447223782539368},{"id":"https://openalex.org/C84211073","wikidata":"https://www.wikidata.org/wiki/Q117879","display_name":"Floating point","level":2,"score":0.5233449339866638},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.4933909475803375},{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.4483276903629303},{"id":"https://openalex.org/C163258240","wikidata":"https://www.wikidata.org/wiki/Q25342","display_name":"Power (physics)","level":2,"score":0.4249325394630432},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.39714333415031433},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.3616896867752075},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3352113962173462},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.32561981678009033},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.26652562618255615},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.09187355637550354},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C6557445","wikidata":"https://www.wikidata.org/wiki/Q173113","display_name":"Agronomy","level":1,"score":0.0},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/tvlsi.2023.3235760","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/tvlsi.2023.3235760","pdf_url":null,"source":{"id":"https://openalex.org/S37538908","display_name":"IEEE Transactions on Very Large Scale Integration (VLSI) Systems","issn_l":"1063-8210","issn":["1063-8210","1557-9999"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Very Large Scale Integration (VLSI) Systems","raw_type":"journal-article"},{"id":"pmh:oai:arXiv.org:2312.09580","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2312.09580","pdf_url":"https://arxiv.org/pdf/2312.09580","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc-sa","license_id":"https://openalex.org/licenses/cc-by-nc-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2312.09580","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2312.09580","pdf_url":"https://arxiv.org/pdf/2312.09580","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc-sa","license_id":"https://openalex.org/licenses/cc-by-nc-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/7","score":0.8999999761581421,"display_name":"Affordable and clean energy"}],"awards":[{"id":"https://openalex.org/G3500348051","display_name":null,"funder_award_id":"111-2622-8-A49-018-SB","funder_id":"https://openalex.org/F4320322795","funder_display_name":"Ministry of Science and Technology, Taiwan"},{"id":"https://openalex.org/G3752499876","display_name":null,"funder_award_id":"110-2218-E-A49-015-MBK.","funder_id":"https://openalex.org/F4320322795","funder_display_name":"Ministry of Science and Technology, Taiwan"},{"id":"https://openalex.org/G4529196702","display_name":null,"funder_award_id":"110-2221-E-A49-148-MY3","funder_id":"https://openalex.org/F4320322795","funder_display_name":"Ministry of Science and Technology, Taiwan"}],"funders":[{"id":"https://openalex.org/F4320322795","display_name":"Ministry of Science and Technology, Taiwan","ror":"https://ror.org/02kv4zf79"},{"id":"https://openalex.org/F4320329659","display_name":"Thailand Science Research and Innovation","ror":null}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4317038458.pdf","grobid_xml":"https://content.openalex.org/works/W4317038458.grobid-xml"},"referenced_works_count":35,"referenced_works":["https://openalex.org/W2289252105","https://openalex.org/W2604319603","https://openalex.org/W2613119772","https://openalex.org/W2809624076","https://openalex.org/W2896409484","https://openalex.org/W2916985722","https://openalex.org/W2938109705","https://openalex.org/W2945146780","https://openalex.org/W2950656546","https://openalex.org/W2952218014","https://openalex.org/W2970821029","https://openalex.org/W2979042679","https://openalex.org/W3000364559","https://openalex.org/W3012561096","https://openalex.org/W3015700067","https://openalex.org/W3021847282","https://openalex.org/W3024621361","https://openalex.org/W3034429256","https://openalex.org/W3036695099","https://openalex.org/W3048902839","https://openalex.org/W3089500695","https://openalex.org/W3099330747","https://openalex.org/W3117297074","https://openalex.org/W3135697241","https://openalex.org/W3164605550","https://openalex.org/W3198609354","https://openalex.org/W4221091807","https://openalex.org/W4240168186","https://openalex.org/W4242765802","https://openalex.org/W4251575795","https://openalex.org/W4292564261","https://openalex.org/W6753069482","https://openalex.org/W6755843862","https://openalex.org/W6767032739","https://openalex.org/W6771680064"],"related_works":["https://openalex.org/W2373300491","https://openalex.org/W4375867731","https://openalex.org/W2378744544","https://openalex.org/W2594301978","https://openalex.org/W2379704676","https://openalex.org/W1998810860","https://openalex.org/W4206442282","https://openalex.org/W4390846322","https://openalex.org/W4280610722","https://openalex.org/W3185228140"],"abstract_inverted_index":{"Low-power":[0],"deep":[1],"learning":[2],"accelerators":[3,20],"(DLAs)":[4],"on":[5,12,27],"the":[6,18,46,49,65,72,79,91,97,100,125,135,149],"speech":[7,38,156],"processing":[8],"enable":[9],"real-time":[10,155],"applications":[11,29],"edge":[13],"devices.":[14],"However,":[15],"most":[16],"of":[17,71,90,124],"existing":[19],"suffer":[21],"from":[22],"high-power":[23],"consumption":[24],"and":[25,42,61,82,93,113,128,140,158,171,177],"focus":[26],"image":[28],"only.":[30],"This":[31],"article":[32],"presents":[33],"a":[34,117],"low-power":[35,129],"accelerator":[36],"for":[37],"separation":[39,157],"through":[40],"algorithm":[41,47],"hardware":[43,98,120],"optimizations.":[44],"At":[45,96],"level,":[48,99],"model":[50,102,136],"is":[51,103],"compressed":[52,101],"with":[53,78,109,116,148],"structured":[54],"sensitivity":[55],"as":[56,58],"well":[57],"unstructured":[59],"pruning,":[60],"further":[62],"quantized":[63],"to":[64,121],"shifted":[66],"8-bit":[67],"floating-point":[68,74],"format":[69],"instead":[70],"32-bit":[73],"format.":[75],"The":[76,131,145,167],"computations":[77],"zero":[80,83],"kernel":[81],"activation":[84,126],"values":[85],"are":[86,174],"skipped":[87],"by":[88,106,138,143],"decomposition":[89],"dilated":[92],"transposed":[94],"convolutions.":[95],"then":[104],"supported":[105],"an":[107],"architecture":[108],"eight":[110],"independent":[111],"multipliers":[112],"accumulators":[114],"(MACs)":[115],"simple":[118],"zero-skipping":[119],"take":[122],"advantage":[123],"sparsity":[127],"processing.":[130],"proposed":[132],"approach":[133],"reduces":[134],"size":[137],"95.44%":[139],"computation":[141],"complexity":[142],"93.88%.":[144],"final":[146],"implementation":[147],"TSMC":[150],"40-nm":[151],"process":[152],"can":[153],"achieve":[154],"consumes":[159],"1.6-mW":[160],"power":[161],"when":[162],"operated":[163],"at":[164],"150":[165],"MHz.":[166],"normalized":[168],"energy":[169],"efficiency":[170,173],"area":[172],"2.344":[175],"TOPS/W":[176],"14.42":[178],"GOPS/mm2,":[179],"respectively.":[180]},"counts_by_year":[{"year":2025,"cited_by_count":3}],"updated_date":"2026-03-20T23:20:44.827607","created_date":"2025-10-10T00:00:00"}