{"id":"https://openalex.org/W3158424777","doi":"https://doi.org/10.1109/icassp43922.2022.9747161","title":"End-to-End Speech Recognition from Federated Acoustic Models","display_name":"End-to-End Speech Recognition from Federated Acoustic Models","publication_year":2022,"publication_date":"2022-04-27","ids":{"openalex":"https://openalex.org/W3158424777","doi":"https://doi.org/10.1109/icassp43922.2022.9747161","mag":"3158424777"},"language":"en","primary_location":{"id":"doi:10.1109/icassp43922.2022.9747161","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp43922.2022.9747161","pdf_url":null,"source":{"id":"https://openalex.org/S4363607702","display_name":"ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","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":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://hal.science/hal-03601224","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100778706","display_name":"Yan Gao","orcid":"https://orcid.org/0000-0002-6852-2038"},"institutions":[{"id":"https://openalex.org/I241749","display_name":"University of Cambridge","ror":"https://ror.org/013meh722","country_code":"GB","type":"education","lineage":["https://openalex.org/I241749"]}],"countries":["GB"],"is_corresponding":true,"raw_author_name":"Yan Gao","raw_affiliation_strings":["University of Cambridge"],"affiliations":[{"raw_affiliation_string":"University of Cambridge","institution_ids":["https://openalex.org/I241749"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5089505434","display_name":"Titouan Parcollet","orcid":"https://orcid.org/0000-0003-0672-1346"},"institutions":[{"id":"https://openalex.org/I241749","display_name":"University of Cambridge","ror":"https://ror.org/013meh722","country_code":"GB","type":"education","lineage":["https://openalex.org/I241749"]},{"id":"https://openalex.org/I198415970","display_name":"Universit\u00e9 d'Avignon et des Pays de Vaucluse","ror":"https://ror.org/00mfpxb84","country_code":"FR","type":"education","lineage":["https://openalex.org/I198415970"]}],"countries":["FR","GB"],"is_corresponding":false,"raw_author_name":"Titouan Parcollet","raw_affiliation_strings":["Avignon University","University of Cambridge"],"affiliations":[{"raw_affiliation_string":"Avignon University","institution_ids":["https://openalex.org/I198415970"]},{"raw_affiliation_string":"University of Cambridge","institution_ids":["https://openalex.org/I241749"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5040954517","display_name":"Salah Zaiem","orcid":null},"institutions":[{"id":"https://openalex.org/I12356871","display_name":"T\u00e9l\u00e9com Paris","ror":"https://ror.org/01naq7912","country_code":"FR","type":"education","lineage":["https://openalex.org/I12356871","https://openalex.org/I205703379","https://openalex.org/I4210145102"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Salah Zaiem","raw_affiliation_strings":["Telecom Paris"],"affiliations":[{"raw_affiliation_string":"Telecom Paris","institution_ids":["https://openalex.org/I12356871"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5031116318","display_name":"Javier Fern\u00e1ndez-Marqu\u00e9s","orcid":null},"institutions":[{"id":"https://openalex.org/I40120149","display_name":"University of Oxford","ror":"https://ror.org/052gg0110","country_code":"GB","type":"education","lineage":["https://openalex.org/I40120149"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Javier Fernandez-Marques","raw_affiliation_strings":["University of Oxford"],"affiliations":[{"raw_affiliation_string":"University of Oxford","institution_ids":["https://openalex.org/I40120149"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5110718216","display_name":"Pedro Porto Buarque de Gusm\u00e3o","orcid":null},"institutions":[{"id":"https://openalex.org/I241749","display_name":"University of Cambridge","ror":"https://ror.org/013meh722","country_code":"GB","type":"education","lineage":["https://openalex.org/I241749"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Pedro P. B. de Gusmao","raw_affiliation_strings":["University of Cambridge"],"affiliations":[{"raw_affiliation_string":"University of Cambridge","institution_ids":["https://openalex.org/I241749"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5075204958","display_name":"Daniel J. Beutel","orcid":"https://orcid.org/0009-0008-6680-5959"},"institutions":[{"id":"https://openalex.org/I241749","display_name":"University of Cambridge","ror":"https://ror.org/013meh722","country_code":"GB","type":"education","lineage":["https://openalex.org/I241749"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Daniel J. Beutel","raw_affiliation_strings":["University of Cambridge","Adap GmbH"],"affiliations":[{"raw_affiliation_string":"University of Cambridge","institution_ids":["https://openalex.org/I241749"]},{"raw_affiliation_string":"Adap GmbH","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5026859250","display_name":"Nicholas D. Lane","orcid":"https://orcid.org/0000-0002-9954-6366"},"institutions":[{"id":"https://openalex.org/I241749","display_name":"University of Cambridge","ror":"https://ror.org/013meh722","country_code":"GB","type":"education","lineage":["https://openalex.org/I241749"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Nicholas D. Lane","raw_affiliation_strings":["University of Cambridge"],"affiliations":[{"raw_affiliation_string":"University of Cambridge","institution_ids":["https://openalex.org/I241749"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5100778706"],"corresponding_institution_ids":["https://openalex.org/I241749"],"apc_list":null,"apc_paid":null,"fwci":3.4443,"has_fulltext":false,"cited_by_count":34,"citation_normalized_percentile":{"value":0.93830498,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":89,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"7227","last_page":"7231"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9998999834060669,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.9998999834060669,"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/T11309","display_name":"Music and Audio Processing","score":0.9993000030517578,"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/T10860","display_name":"Speech and Audio Processing","score":0.9979000091552734,"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":0.827578604221344},{"id":"https://openalex.org/keywords/weighting","display_name":"Weighting","score":0.6108422875404358},{"id":"https://openalex.org/keywords/sequence","display_name":"Sequence (biology)","score":0.5471779108047485},{"id":"https://openalex.org/keywords/acoustic-model","display_name":"Acoustic model","score":0.5439155101776123},{"id":"https://openalex.org/keywords/construct","display_name":"Construct (python library)","score":0.5217231512069702},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.510543942451477},{"id":"https://openalex.org/keywords/end-to-end-principle","display_name":"End-to-end principle","score":0.4694487750530243},{"id":"https://openalex.org/keywords/word-error-rate","display_name":"Word error rate","score":0.46757179498672485},{"id":"https://openalex.org/keywords/word","display_name":"Word (group theory)","score":0.44866687059402466},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.37907126545906067},{"id":"https://openalex.org/keywords/speech-processing","display_name":"Speech processing","score":0.20711681246757507},{"id":"https://openalex.org/keywords/acoustics","display_name":"Acoustics","score":0.08514779806137085}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.827578604221344},{"id":"https://openalex.org/C183115368","wikidata":"https://www.wikidata.org/wiki/Q856577","display_name":"Weighting","level":2,"score":0.6108422875404358},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.5471779108047485},{"id":"https://openalex.org/C155635449","wikidata":"https://www.wikidata.org/wiki/Q4674699","display_name":"Acoustic model","level":3,"score":0.5439155101776123},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.5217231512069702},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.510543942451477},{"id":"https://openalex.org/C74296488","wikidata":"https://www.wikidata.org/wiki/Q2527392","display_name":"End-to-end principle","level":2,"score":0.4694487750530243},{"id":"https://openalex.org/C40969351","wikidata":"https://www.wikidata.org/wiki/Q3516228","display_name":"Word error rate","level":2,"score":0.46757179498672485},{"id":"https://openalex.org/C90805587","wikidata":"https://www.wikidata.org/wiki/Q10944557","display_name":"Word (group theory)","level":2,"score":0.44866687059402466},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.37907126545906067},{"id":"https://openalex.org/C61328038","wikidata":"https://www.wikidata.org/wiki/Q3358061","display_name":"Speech processing","level":2,"score":0.20711681246757507},{"id":"https://openalex.org/C24890656","wikidata":"https://www.wikidata.org/wiki/Q82811","display_name":"Acoustics","level":1,"score":0.08514779806137085},{"id":"https://openalex.org/C54355233","wikidata":"https://www.wikidata.org/wiki/Q7162","display_name":"Genetics","level":1,"score":0.0},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","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/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.0},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/icassp43922.2022.9747161","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp43922.2022.9747161","pdf_url":null,"source":{"id":"https://openalex.org/S4363607702","display_name":"ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","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":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},{"id":"pmh:oai:HAL:hal-03601224v1","is_oa":true,"landing_page_url":"https://hal.science/hal-03601224","pdf_url":null,"source":{"id":"https://openalex.org/S4306402512","display_name":"HAL (Le Centre pour la Communication Scientifique Directe)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1294671590","host_organization_name":"Centre National de la Recherche Scientifique","host_organization_lineage":["https://openalex.org/I1294671590"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"The International Conference on Acoustics, Speech, & Signal Processing (ICASSP), May 2022, Singapour, Singapore","raw_type":"Conference papers"}],"best_oa_location":{"id":"pmh:oai:HAL:hal-03601224v1","is_oa":true,"landing_page_url":"https://hal.science/hal-03601224","pdf_url":null,"source":{"id":"https://openalex.org/S4306402512","display_name":"HAL (Le Centre pour la Communication Scientifique Directe)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1294671590","host_organization_name":"Centre National de la Recherche Scientifique","host_organization_lineage":["https://openalex.org/I1294671590"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"The International Conference on Acoustics, Speech, & Signal Processing (ICASSP), May 2022, Singapour, Singapore","raw_type":"Conference papers"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":66,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W1922655562","https://openalex.org/W2046932483","https://openalex.org/W2102113734","https://openalex.org/W2118774185","https://openalex.org/W2193413348","https://openalex.org/W2526425061","https://openalex.org/W2535838896","https://openalex.org/W2541884796","https://openalex.org/W2545177271","https://openalex.org/W2697044473","https://openalex.org/W2739427748","https://openalex.org/W2807006176","https://openalex.org/W2896422817","https://openalex.org/W2914328083","https://openalex.org/W2941814890","https://openalex.org/W2951327905","https://openalex.org/W2954045886","https://openalex.org/W2962824709","https://openalex.org/W2962826786","https://openalex.org/W2963702081","https://openalex.org/W2963827914","https://openalex.org/W2964172053","https://openalex.org/W2985913104","https://openalex.org/W2989289980","https://openalex.org/W2995929068","https://openalex.org/W3008187686","https://openalex.org/W3008191852","https://openalex.org/W3015585292","https://openalex.org/W3015783745","https://openalex.org/W3021654819","https://openalex.org/W3025567392","https://openalex.org/W3026715795","https://openalex.org/W3030437843","https://openalex.org/W3046183292","https://openalex.org/W3047081942","https://openalex.org/W3093436911","https://openalex.org/W3094793624","https://openalex.org/W3094992563","https://openalex.org/W3095866088","https://openalex.org/W3097714942","https://openalex.org/W3103802018","https://openalex.org/W3127184075","https://openalex.org/W3134509799","https://openalex.org/W3162072286","https://openalex.org/W3167533889","https://openalex.org/W3196787717","https://openalex.org/W4287322665","https://openalex.org/W4294106961","https://openalex.org/W4297060071","https://openalex.org/W4297687186","https://openalex.org/W4299283926","https://openalex.org/W4318619660","https://openalex.org/W6640090968","https://openalex.org/W6687566353","https://openalex.org/W6728757088","https://openalex.org/W6728910023","https://openalex.org/W6752029299","https://openalex.org/W6761563299","https://openalex.org/W6770434673","https://openalex.org/W6771467084","https://openalex.org/W6771536673","https://openalex.org/W6773976177","https://openalex.org/W6781175393","https://openalex.org/W6784215088","https://openalex.org/W6791347023"],"related_works":["https://openalex.org/W2944691285","https://openalex.org/W3081187864","https://openalex.org/W151018310","https://openalex.org/W4380605396","https://openalex.org/W2916997151","https://openalex.org/W3198455051","https://openalex.org/W3133352777","https://openalex.org/W2949174760","https://openalex.org/W2008737763","https://openalex.org/W2584084702"],"abstract_inverted_index":{"Training":[0],"Automatic":[1],"Speech":[2],"Recognition":[3],"(ASR)":[4],"models":[5,153],"under":[6],"federated":[7,49,151],"learning":[8],"(FL)":[9],"settings":[10],"has":[11],"attracted":[12],"a":[13,44,69,107],"lot":[14],"of":[15,35,53,65,75,162],"attention":[16],"recently.":[17],"However,":[18],"the":[19,25,33,60,66,84,137,155],"FL":[20,37,118,164],"scenarios":[21],"often":[22],"presented":[23],"in":[24,115],"literature":[26],"are":[27],"artificial":[28],"and":[29,46,62,80,106,124,128,149,160],"fail":[30],"to":[31],"capture":[32],"complexity":[34],"real":[36],"systems.":[38],"In":[39],"this":[40],"paper,":[41],"we":[42],"construct":[43],"challenging":[45],"realistic":[47,117,150,163],"ASR":[48,94,134,146,165],"experimental":[50],"setup":[51],"consisting":[52],"clients":[54,123],"with":[55,96,121,126],"heterogeneous":[56,71,148],"data":[57],"distributions":[58],"using":[59],"French":[61],"Italian":[63],"sets":[64],"CommonVoice":[67],"dataset,":[68],"large":[70],"dataset":[72],"containing":[73],"thousands":[74],"different":[76],"speakers,":[77],"acoustic":[78,152],"environments":[79],"noises.":[81],"We":[82],"present":[83],"first":[85],"empirical":[86],"study":[87],"on":[88,144],"an":[89],"attention-based":[90],"sequence-to-sequence":[91],"End-to-End":[92],"(E2E)":[93],"model":[95],"three":[97],"aggregation":[98,105],"weighting":[99],"strategies":[100],"\u2013":[101],"standard":[102],"FedAvg,":[103],"loss-based":[104],"novel":[108],"word":[109],"error":[110],"rate":[111],"(WER)-based":[112],"aggregation,":[113],"compared":[114],"two":[116],"scenarios:":[119],"cross-silo":[120],"10":[122],"cross-device":[125,133],"2K":[127],"4K":[129,132],"clients.":[130],"This":[131],"experiment":[135],"is":[136],"largest":[138],"ever":[139],"performed.":[140],"Our":[141],"first-of-its-kind":[142],"analysis":[143],"E2E":[145],"from":[147],"provides":[154],"foundations":[156],"for":[157],"future":[158],"research":[159],"development":[161],"applications.":[166]},"counts_by_year":[{"year":2025,"cited_by_count":5},{"year":2024,"cited_by_count":8},{"year":2023,"cited_by_count":14},{"year":2022,"cited_by_count":6},{"year":2021,"cited_by_count":1}],"updated_date":"2026-04-05T17:49:38.594831","created_date":"2025-10-10T00:00:00"}