{"id":"https://openalex.org/W4400524955","doi":"https://doi.org/10.1109/access.2024.3426471","title":"Benchmarking Inference of Transformer-Based Transcription Models With Clustering on Embedded GPUs","display_name":"Benchmarking Inference of Transformer-Based Transcription Models With Clustering on Embedded GPUs","publication_year":2024,"publication_date":"2024-01-01","ids":{"openalex":"https://openalex.org/W4400524955","doi":"https://doi.org/10.1109/access.2024.3426471"},"language":"en","primary_location":{"id":"doi:10.1109/access.2024.3426471","is_oa":true,"landing_page_url":"https://doi.org/10.1109/access.2024.3426471","pdf_url":null,"source":{"id":"https://openalex.org/S2485537415","display_name":"IEEE Access","issn_l":"2169-3536","issn":["2169-3536"],"is_oa":true,"is_in_doaj":true,"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Access","raw_type":"journal-article"},"type":"article","indexed_in":["crossref","doaj"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1109/access.2024.3426471","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5102920725","display_name":"Marika E. Schubert","orcid":"https://orcid.org/0000-0002-9270-8502"},"institutions":[{"id":"https://openalex.org/I170201317","display_name":"University of Pittsburgh","ror":"https://ror.org/01an3r305","country_code":"US","type":"education","lineage":["https://openalex.org/I170201317"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Marika E. Schubert","raw_affiliation_strings":["Department of Electrical and Computer Engineering, University of Pittsburgh, Pittsburgh, PA, USA"],"raw_orcid":"https://orcid.org/0000-0002-9270-8502","affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, University of Pittsburgh, Pittsburgh, PA, USA","institution_ids":["https://openalex.org/I170201317"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5042589258","display_name":"David Langerman","orcid":"https://orcid.org/0000-0001-8777-4655"},"institutions":[{"id":"https://openalex.org/I170201317","display_name":"University of Pittsburgh","ror":"https://ror.org/01an3r305","country_code":"US","type":"education","lineage":["https://openalex.org/I170201317"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"David Langerman","raw_affiliation_strings":["NSF Center of Space, High-Performance, and Resilient Computing, University of Pittsburgh, Pittsburgh, PA, USA","NSF Center of Space, High-Performance, and Resilient Computing at the University of Pittsburgh, Pittsburgh, PA, USA"],"raw_orcid":"https://orcid.org/0000-0001-8777-4655","affiliations":[{"raw_affiliation_string":"NSF Center of Space, High-Performance, and Resilient Computing, University of Pittsburgh, Pittsburgh, PA, USA","institution_ids":["https://openalex.org/I170201317"]},{"raw_affiliation_string":"NSF Center of Space, High-Performance, and Resilient Computing at the University of Pittsburgh, Pittsburgh, PA, USA","institution_ids":["https://openalex.org/I170201317"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5082898376","display_name":"Alan D. George","orcid":"https://orcid.org/0000-0001-9665-2879"},"institutions":[{"id":"https://openalex.org/I170201317","display_name":"University of Pittsburgh","ror":"https://ror.org/01an3r305","country_code":"US","type":"education","lineage":["https://openalex.org/I170201317"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Alan D. George","raw_affiliation_strings":["Department of Electrical and Computer Engineering, University of Pittsburgh, Pittsburgh, PA, USA"],"raw_orcid":"https://orcid.org/0000-0001-9665-2879","affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, University of Pittsburgh, Pittsburgh, PA, USA","institution_ids":["https://openalex.org/I170201317"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I170201317"],"apc_list":{"value":1850,"currency":"USD","value_usd":1850},"apc_paid":{"value":1850,"currency":"USD","value_usd":1850},"fwci":0.5972,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.72384474,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":95,"max":96},"biblio":{"volume":"12","issue":null,"first_page":"123276","last_page":"123293"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9973000288009644,"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"}},"topics":[{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9973000288009644,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.9957000017166138,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9948999881744385,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.8610981702804565},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8309648633003235},{"id":"https://openalex.org/keywords/benchmarking","display_name":"Benchmarking","score":0.7380688190460205},{"id":"https://openalex.org/keywords/bottleneck","display_name":"Bottleneck","score":0.7222139239311218},{"id":"https://openalex.org/keywords/cluster-analysis","display_name":"Cluster analysis","score":0.6059946417808533},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.5630720257759094},{"id":"https://openalex.org/keywords/energy-consumption","display_name":"Energy consumption","score":0.5221213102340698},{"id":"https://openalex.org/keywords/software-deployment","display_name":"Software deployment","score":0.490339994430542},{"id":"https://openalex.org/keywords/efficient-energy-use","display_name":"Efficient energy use","score":0.44365739822387695},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.35583698749542236},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3085256516933441},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.22815027832984924},{"id":"https://openalex.org/keywords/voltage","display_name":"Voltage","score":0.09392645955085754}],"concepts":[{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.8610981702804565},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8309648633003235},{"id":"https://openalex.org/C86251818","wikidata":"https://www.wikidata.org/wiki/Q816754","display_name":"Benchmarking","level":2,"score":0.7380688190460205},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.7222139239311218},{"id":"https://openalex.org/C73555534","wikidata":"https://www.wikidata.org/wiki/Q622825","display_name":"Cluster analysis","level":2,"score":0.6059946417808533},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.5630720257759094},{"id":"https://openalex.org/C2780165032","wikidata":"https://www.wikidata.org/wiki/Q16869822","display_name":"Energy consumption","level":2,"score":0.5221213102340698},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.490339994430542},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.44365739822387695},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.35583698749542236},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3085256516933441},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.22815027832984924},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.09392645955085754},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","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/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C144133560","wikidata":"https://www.wikidata.org/wiki/Q4830453","display_name":"Business","level":0,"score":0.0},{"id":"https://openalex.org/C162853370","wikidata":"https://www.wikidata.org/wiki/Q39809","display_name":"Marketing","level":1,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.0},{"id":"https://openalex.org/C18903297","wikidata":"https://www.wikidata.org/wiki/Q7150","display_name":"Ecology","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/access.2024.3426471","is_oa":true,"landing_page_url":"https://doi.org/10.1109/access.2024.3426471","pdf_url":null,"source":{"id":"https://openalex.org/S2485537415","display_name":"IEEE Access","issn_l":"2169-3536","issn":["2169-3536"],"is_oa":true,"is_in_doaj":true,"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Access","raw_type":"journal-article"},{"id":"pmh:oai:doaj.org/article:9b3a3f1564d7460c894ab6ce1fb3819d","is_oa":true,"landing_page_url":"https://doaj.org/article/9b3a3f1564d7460c894ab6ce1fb3819d","pdf_url":null,"source":{"id":"https://openalex.org/S4306401280","display_name":"DOAJ (DOAJ: Directory of Open Access Journals)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"IEEE Access, Vol 12, Pp 123276-123293 (2024)","raw_type":"article"}],"best_oa_location":{"id":"doi:10.1109/access.2024.3426471","is_oa":true,"landing_page_url":"https://doi.org/10.1109/access.2024.3426471","pdf_url":null,"source":{"id":"https://openalex.org/S2485537415","display_name":"IEEE Access","issn_l":"2169-3536","issn":["2169-3536"],"is_oa":true,"is_in_doaj":true,"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Access","raw_type":"journal-article"},"sustainable_development_goals":[{"display_name":"Affordable and clean energy","score":0.4699999988079071,"id":"https://metadata.un.org/sdg/7"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":40,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W1922655562","https://openalex.org/W2127141656","https://openalex.org/W2169818249","https://openalex.org/W2901739041","https://openalex.org/W2933138175","https://openalex.org/W2962760690","https://openalex.org/W2963266252","https://openalex.org/W2964045208","https://openalex.org/W2965373594","https://openalex.org/W2973049979","https://openalex.org/W2996428491","https://openalex.org/W3035281298","https://openalex.org/W3037798801","https://openalex.org/W3093579165","https://openalex.org/W3100110884","https://openalex.org/W3160525311","https://openalex.org/W3167869698","https://openalex.org/W3209059054","https://openalex.org/W4205779689","https://openalex.org/W4287725215","https://openalex.org/W4311457721","https://openalex.org/W4322718253","https://openalex.org/W4385245566","https://openalex.org/W6631362777","https://openalex.org/W6640090968","https://openalex.org/W6687566353","https://openalex.org/W6755207826","https://openalex.org/W6756104738","https://openalex.org/W6756319913","https://openalex.org/W6768021236","https://openalex.org/W6769196770","https://openalex.org/W6771467084","https://openalex.org/W6779709467","https://openalex.org/W6780218876","https://openalex.org/W6780333223","https://openalex.org/W6784614252","https://openalex.org/W6795952400","https://openalex.org/W6810007534","https://openalex.org/W6850162387"],"related_works":["https://openalex.org/W4238897586","https://openalex.org/W435179959","https://openalex.org/W2619091065","https://openalex.org/W2059640416","https://openalex.org/W1490753184","https://openalex.org/W2284465472","https://openalex.org/W2291782699","https://openalex.org/W1993948687","https://openalex.org/W2000169967","https://openalex.org/W2112883198"],"abstract_inverted_index":{"Early":[0],"awareness":[1],"of":[2,8,81,96,108,210,240],"inference":[3,27,36,74,94,201,235],"performance":[4,23,37,71,95,236],"ensures":[5],"the":[6,79,93,97,123,180,196,215,225],"feasibility":[7],"machine":[9],"learning":[10],"for":[11,62,175,185,202],"embedded":[12,64],"deployment.":[13],"Often,":[14],"ML":[15],"model":[16,35,47,100,110,161,172,182,194,217,248],"selection":[17],"often":[18,39],"focuses":[19],"first":[20],"on":[21,92,113],"training":[22,32,84],"and":[24,88,105,131,166,237],"accuracy,":[25],"with":[26,134],"considered":[28],"second.":[29],"While":[30],"prioritizing":[31],"is":[33,38,60,111,120],"necessary,":[34],"impacted":[40],"by":[41],"these":[42,229],"optimizations.":[43,249],"Knowing":[44],"whether":[45],"a":[46,137,146],"will":[48],"run":[49],"meaningfully":[50],"faster":[51,174,184],"or":[52],"be":[53,152,244],"more":[54,220],"energy":[55,106,199,238],"efficient":[56],"than":[57,189,206,224],"its":[58,90],"un-optimized-counterpart":[59],"required":[61],"resource-constrained":[63],"environments.":[65],"Training-time":[66],"optimizations":[67],"may":[68],"incur":[69],"real":[70],"losses":[72],"at":[73],"time.":[75],"This":[76],"paper":[77],"benchmarks":[78],"effect":[80,91],"one":[82],"such":[83],"optimization,":[85],"clustered":[86,181,193],"attention,":[87],"examines":[89],"transformer-based":[98],"transcription":[99],"wav2vec2.":[101],"The":[102,170,192],"execution":[103],"time":[104,221],"consumption":[107,239],"this":[109],"evaluated":[112],"NVIDIA":[114],"Jetson-embedded":[115],"GPU":[116],"devices.":[117],"Clustered":[118],"attention":[119,168],"specific":[121],"to":[122,136,222],"transformer":[124],"self-attention":[125],"mechanism,":[126],"which":[127],"exhibits":[128],"poor":[129],"memory":[130],"execution-time":[132],"scaling":[133,142],"respect":[135],"variable":[138],"input":[139,177,186,203],"size.":[140],"These":[141],"characteristics":[143],"make":[144],"it":[145],"potentially":[147],"critical":[148],"bottleneck":[149],"that":[150,234],"must":[151],"observed":[153],"under":[154],"realistic":[155],"conditions.":[156],"Our":[157],"research":[158],"considers":[159],"three":[160],"variants:":[162],"reference":[163,171,226],"(original),":[164],"clustered,":[165],"improved-clustered":[167,216],"models.":[169],"was":[173,183],"small":[176],"sizes,":[178],"but":[179],"sizes":[187,204],"longer":[188],"10":[190],"seconds.":[191],"had":[195],"lowest":[197],"maximum":[198],"per":[200],"greater":[205],"about":[207],"12":[208],"seconds":[209],"audio.":[211],"With":[212,228],"optimal":[213],"configuration,":[214],"takes":[218],"26.34%":[219],"execute":[223],"model.":[227],"operational":[230],"differences,":[231],"we":[232],"show":[233],"deployment":[241],"should":[242],"not":[243],"overlooked":[245],"in":[246],"selecting":[247]},"counts_by_year":[{"year":2025,"cited_by_count":2}],"updated_date":"2026-06-26T08:34:08.712188","created_date":"2025-10-10T00:00:00"}