{"id":"https://openalex.org/W3206876927","doi":"https://doi.org/10.1109/icassp43922.2022.9746814","title":"Efficient Sequence Training of Attention Models Using Approximative Recombination","display_name":"Efficient Sequence Training of Attention Models Using Approximative Recombination","publication_year":2022,"publication_date":"2022-04-27","ids":{"openalex":"https://openalex.org/W3206876927","doi":"https://doi.org/10.1109/icassp43922.2022.9746814","mag":"3206876927"},"language":"en","primary_location":{"id":"doi:10.1109/icassp43922.2022.9746814","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp43922.2022.9746814","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":"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/A5029918053","display_name":"Nils-Philipp Wynands","orcid":null},"institutions":[{"id":"https://openalex.org/I887968799","display_name":"RWTH Aachen University","ror":"https://ror.org/04xfq0f34","country_code":"DE","type":"education","lineage":["https://openalex.org/I887968799"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Nils-Philipp Wynands","raw_affiliation_strings":["RWTH Aachen University,Human Language Technology and Pattern Recognition,Computer Science Department,Aachen,Germany,52062"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"RWTH Aachen University,Human Language Technology and Pattern Recognition,Computer Science Department,Aachen,Germany,52062","institution_ids":["https://openalex.org/I887968799"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5029547617","display_name":"Wilfried Michel","orcid":null},"institutions":[{"id":"https://openalex.org/I887968799","display_name":"RWTH Aachen University","ror":"https://ror.org/04xfq0f34","country_code":"DE","type":"education","lineage":["https://openalex.org/I887968799"]},{"id":"https://openalex.org/I99977706","display_name":"FH Aachen","ror":"https://ror.org/04tqgg260","country_code":"DE","type":"education","lineage":["https://openalex.org/I99977706"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Wilfried Michel","raw_affiliation_strings":["RWTH Aachen University,Human Language Technology and Pattern Recognition,Computer Science Department,Aachen,Germany,52062","AppTek GmbH, Aachen, Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"RWTH Aachen University,Human Language Technology and Pattern Recognition,Computer Science Department,Aachen,Germany,52062","institution_ids":["https://openalex.org/I887968799"]},{"raw_affiliation_string":"AppTek GmbH, Aachen, Germany","institution_ids":["https://openalex.org/I99977706"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5082142903","display_name":"Jan Rosendahl","orcid":null},"institutions":[{"id":"https://openalex.org/I887968799","display_name":"RWTH Aachen University","ror":"https://ror.org/04xfq0f34","country_code":"DE","type":"education","lineage":["https://openalex.org/I887968799"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Jan Rosendahl","raw_affiliation_strings":["RWTH Aachen University,Human Language Technology and Pattern Recognition,Computer Science Department,Aachen,Germany,52062"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"RWTH Aachen University,Human Language Technology and Pattern Recognition,Computer Science Department,Aachen,Germany,52062","institution_ids":["https://openalex.org/I887968799"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5088968292","display_name":"Ralf Schl\u00fcter","orcid":"https://orcid.org/0000-0003-2839-9247"},"institutions":[{"id":"https://openalex.org/I887968799","display_name":"RWTH Aachen University","ror":"https://ror.org/04xfq0f34","country_code":"DE","type":"education","lineage":["https://openalex.org/I887968799"]},{"id":"https://openalex.org/I99977706","display_name":"FH Aachen","ror":"https://ror.org/04tqgg260","country_code":"DE","type":"education","lineage":["https://openalex.org/I99977706"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Ralf Schluter","raw_affiliation_strings":["RWTH Aachen University,Human Language Technology and Pattern Recognition,Computer Science Department,Aachen,Germany,52062","AppTek GmbH, Aachen, Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"RWTH Aachen University,Human Language Technology and Pattern Recognition,Computer Science Department,Aachen,Germany,52062","institution_ids":["https://openalex.org/I887968799"]},{"raw_affiliation_string":"AppTek GmbH, Aachen, Germany","institution_ids":["https://openalex.org/I99977706"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5112501010","display_name":"Hermann Ney","orcid":null},"institutions":[{"id":"https://openalex.org/I887968799","display_name":"RWTH Aachen University","ror":"https://ror.org/04xfq0f34","country_code":"DE","type":"education","lineage":["https://openalex.org/I887968799"]},{"id":"https://openalex.org/I99977706","display_name":"FH Aachen","ror":"https://ror.org/04tqgg260","country_code":"DE","type":"education","lineage":["https://openalex.org/I99977706"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Hermann Ney","raw_affiliation_strings":["RWTH Aachen University,Human Language Technology and Pattern Recognition,Computer Science Department,Aachen,Germany,52062","AppTek GmbH, Aachen, Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"RWTH Aachen University,Human Language Technology and Pattern Recognition,Computer Science Department,Aachen,Germany,52062","institution_ids":["https://openalex.org/I887968799"]},{"raw_affiliation_string":"AppTek GmbH, Aachen, Germany","institution_ids":["https://openalex.org/I99977706"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.5061,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.60601728,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":96},"biblio":{"volume":"45","issue":null,"first_page":"8002","last_page":"8006"},"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/T10860","display_name":"Speech and Audio Processing","score":0.9995999932289124,"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/T11309","display_name":"Music and Audio Processing","score":0.9972000122070312,"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/discriminative-model","display_name":"Discriminative model","score":0.852734386920929},{"id":"https://openalex.org/keywords/sequence","display_name":"Sequence (biology)","score":0.7260677814483643},{"id":"https://openalex.org/keywords/beam-search","display_name":"Beam search","score":0.6987857222557068},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.6709696650505066},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6530642509460449},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.6490270495414734},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.5989329218864441},{"id":"https://openalex.org/keywords/word","display_name":"Word (group theory)","score":0.5492159128189087},{"id":"https://openalex.org/keywords/encoding","display_name":"Encoding (memory)","score":0.49351298809051514},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.46897611021995544},{"id":"https://openalex.org/keywords/limiting","display_name":"Limiting","score":0.4608876407146454},{"id":"https://openalex.org/keywords/word-error-rate","display_name":"Word error rate","score":0.4384719431400299},{"id":"https://openalex.org/keywords/context-model","display_name":"Context model","score":0.4234030842781067},{"id":"https://openalex.org/keywords/state","display_name":"State (computer science)","score":0.4203365445137024},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.40370896458625793},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4017524719238281},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.38432663679122925},{"id":"https://openalex.org/keywords/search-algorithm","display_name":"Search algorithm","score":0.2565375566482544},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.21061217784881592},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.08905625343322754}],"concepts":[{"id":"https://openalex.org/C97931131","wikidata":"https://www.wikidata.org/wiki/Q5282087","display_name":"Discriminative model","level":2,"score":0.852734386920929},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.7260677814483643},{"id":"https://openalex.org/C19889080","wikidata":"https://www.wikidata.org/wiki/Q2835852","display_name":"Beam search","level":3,"score":0.6987857222557068},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.6709696650505066},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6530642509460449},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.6490270495414734},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.5989329218864441},{"id":"https://openalex.org/C90805587","wikidata":"https://www.wikidata.org/wiki/Q10944557","display_name":"Word (group theory)","level":2,"score":0.5492159128189087},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.49351298809051514},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.46897611021995544},{"id":"https://openalex.org/C188198153","wikidata":"https://www.wikidata.org/wiki/Q1613840","display_name":"Limiting","level":2,"score":0.4608876407146454},{"id":"https://openalex.org/C40969351","wikidata":"https://www.wikidata.org/wiki/Q3516228","display_name":"Word error rate","level":2,"score":0.4384719431400299},{"id":"https://openalex.org/C183322885","wikidata":"https://www.wikidata.org/wiki/Q17007702","display_name":"Context model","level":3,"score":0.4234030842781067},{"id":"https://openalex.org/C48103436","wikidata":"https://www.wikidata.org/wiki/Q599031","display_name":"State (computer science)","level":2,"score":0.4203365445137024},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.40370896458625793},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4017524719238281},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.38432663679122925},{"id":"https://openalex.org/C125583679","wikidata":"https://www.wikidata.org/wiki/Q755673","display_name":"Search algorithm","level":2,"score":0.2565375566482544},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.21061217784881592},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.08905625343322754},{"id":"https://openalex.org/C54355233","wikidata":"https://www.wikidata.org/wiki/Q7162","display_name":"Genetics","level":1,"score":0.0},{"id":"https://openalex.org/C201995342","wikidata":"https://www.wikidata.org/wiki/Q682496","display_name":"Systems engineering","level":1,"score":0.0},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0},{"id":"https://openalex.org/C78519656","wikidata":"https://www.wikidata.org/wiki/Q101333","display_name":"Mechanical engineering","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/C2781238097","wikidata":"https://www.wikidata.org/wiki/Q175026","display_name":"Object (grammar)","level":2,"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/C151730666","wikidata":"https://www.wikidata.org/wiki/Q7205","display_name":"Paleontology","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/icassp43922.2022.9746814","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp43922.2022.9746814","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:publications.rwth-aachen.de:888857","is_oa":false,"landing_page_url":"https://publications.rwth-aachen.de/record/888857","pdf_url":null,"source":{"id":"https://openalex.org/S4306401033","display_name":"RWTH Publications (RWTH Aachen)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I887968799","host_organization_name":"RWTH Aachen University","host_organization_lineage":["https://openalex.org/I887968799"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2022 IEEE International Conference on Acoustics, Speech, and Signal Processing : proceedings : 7-13 May 2022, virtual (all paper presentations) : 22-27 May 2022, main venue: Marina Bay Sands Expo & Convention Center, Singapore, satellite venue: Shenzhen, China / sponsored by: the Institute of Electrical and Electronics Engineers, Signal Processing Society<br/>IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2022, online, 2022-05-23 - 2022-05-27","raw_type":"info:eu-repo/semantics/conferenceObject"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/10","display_name":"Reduced inequalities","score":0.7599999904632568}],"awards":[],"funders":[{"id":"https://openalex.org/F4320334678","display_name":"European Research Council","ror":"https://ror.org/0472cxd90"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":15,"referenced_works":["https://openalex.org/W1915251500","https://openalex.org/W2114016253","https://openalex.org/W2902786662","https://openalex.org/W2936774411","https://openalex.org/W2962728618","https://openalex.org/W2963747784","https://openalex.org/W2972953886","https://openalex.org/W3008191852","https://openalex.org/W3036601975","https://openalex.org/W3095376166","https://openalex.org/W3095697114","https://openalex.org/W3099782249","https://openalex.org/W3185108477","https://openalex.org/W3198439131","https://openalex.org/W6780218876"],"related_works":["https://openalex.org/W2965546495","https://openalex.org/W4389116644","https://openalex.org/W2153315159","https://openalex.org/W3103844505","https://openalex.org/W4205463238","https://openalex.org/W259157601","https://openalex.org/W2162582511","https://openalex.org/W1985979081","https://openalex.org/W2145850538","https://openalex.org/W1795026554"],"abstract_inverted_index":{"Sequence":[0],"discriminative":[1,129],"training":[2,130],"is":[3,29,82,87,91,117],"a":[4,21,75],"great":[5],"tool":[6],"to":[7,31,49,63,125],"improve":[8],"the":[9,47,85,97,112,137],"performance":[10],"of":[11,53,67,107],"an":[12,50],"automatic":[13],"speech":[14],"recognition":[15],"system.":[16],"It":[17],"does,":[18],"however,":[19],"necessitate":[20],"sum":[22],"over":[23],"all":[24],"possible":[25],"word":[26],"sequences,":[27],"which":[28],"intractable":[30],"compute":[32],"in":[33],"practice.":[34],"Current":[35],"state-of-the-art":[36],"systems":[37],"with":[38],"unlimited":[39],"label":[40],"context":[41],"circumvent":[42],"this":[43,95,120],"problem":[44],"by":[45,84,104],"limiting":[46],"summation":[48],"n-best":[51],"list":[52],"relevant":[54],"competing":[55],"hypotheses":[56,68],"obtained":[57],"from":[58],"beam":[59,70,99],"search.This":[60],"work":[61],"proposes":[62],"perform":[64,127],"(approximative)":[65],"recombinations":[66],"during":[69],"search,":[71],"if":[72],"they":[73],"share":[74],"common":[76],"local":[77],"history.":[78],"The":[79],"error":[80],"that":[81,93,119],"incurred":[83],"approximation":[86],"analyzed":[88],"and":[89],"it":[90,116],"shown":[92,118],"using":[94],"technique":[96,121],"effective":[98],"size":[100],"can":[101,122],"be":[102,123],"increased":[103],"several":[105],"orders":[106],"magnitude":[108],"without":[109],"significantly":[110],"increasing":[111],"computational":[113],"requirements.":[114],"Lastly,":[115],"used":[124],"effectively":[126],"sequence":[128],"for":[131],"attention-based":[132],"encoder-decoder":[133],"acoustic":[134],"models":[135],"on":[136],"LibriSpeech":[138],"task.":[139]},"counts_by_year":[{"year":2023,"cited_by_count":2},{"year":2022,"cited_by_count":1}],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}