{"id":"https://openalex.org/W3170968215","doi":"https://doi.org/10.21437/interspeech.2021-797","title":"CoDERT: Distilling Encoder Representations with Co-Learning for Transducer-Based Speech Recognition","display_name":"CoDERT: Distilling Encoder Representations with Co-Learning for Transducer-Based Speech Recognition","publication_year":2021,"publication_date":"2021-08-27","ids":{"openalex":"https://openalex.org/W3170968215","doi":"https://doi.org/10.21437/interspeech.2021-797","mag":"3170968215"},"language":"en","primary_location":{"id":"doi:10.21437/interspeech.2021-797","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2021-797","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2021","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2106.07734","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5076547892","display_name":"Rupak Vignesh Swaminathan","orcid":null},"institutions":[{"id":"https://openalex.org/I4210089985","display_name":"Amazon (Germany)","ror":"https://ror.org/00b9ktm87","country_code":"DE","type":"company","lineage":["https://openalex.org/I1311688040","https://openalex.org/I4210089985"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Rupak Vignesh Swaminathan","raw_affiliation_strings":["Amazon"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Amazon","institution_ids":["https://openalex.org/I4210089985"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5027102522","display_name":"Brian King","orcid":"https://orcid.org/0000-0002-5300-5564"},"institutions":[{"id":"https://openalex.org/I4210089985","display_name":"Amazon (Germany)","ror":"https://ror.org/00b9ktm87","country_code":"DE","type":"company","lineage":["https://openalex.org/I1311688040","https://openalex.org/I4210089985"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Brian King","raw_affiliation_strings":["Amazon"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Amazon","institution_ids":["https://openalex.org/I4210089985"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5023708721","display_name":"Grant P. Strimel","orcid":null},"institutions":[{"id":"https://openalex.org/I4210089985","display_name":"Amazon (Germany)","ror":"https://ror.org/00b9ktm87","country_code":"DE","type":"company","lineage":["https://openalex.org/I1311688040","https://openalex.org/I4210089985"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Grant P. Strimel","raw_affiliation_strings":["Amazon"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Amazon","institution_ids":["https://openalex.org/I4210089985"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5012153296","display_name":"Jasha Droppo","orcid":"https://orcid.org/0000-0001-6097-0090"},"institutions":[{"id":"https://openalex.org/I4210089985","display_name":"Amazon (Germany)","ror":"https://ror.org/00b9ktm87","country_code":"DE","type":"company","lineage":["https://openalex.org/I1311688040","https://openalex.org/I4210089985"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Jasha Droppo","raw_affiliation_strings":["Amazon"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Amazon","institution_ids":["https://openalex.org/I4210089985"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5029573067","display_name":"Athanasios Mouchtaris","orcid":"https://orcid.org/0000-0001-7583-0189"},"institutions":[{"id":"https://openalex.org/I4210089985","display_name":"Amazon (Germany)","ror":"https://ror.org/00b9ktm87","country_code":"DE","type":"company","lineage":["https://openalex.org/I1311688040","https://openalex.org/I4210089985"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Athanasios Mouchtaris","raw_affiliation_strings":["Amazon"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Amazon","institution_ids":["https://openalex.org/I4210089985"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I4210089985"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":3,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"4543","last_page":"4547"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":1.0,"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":1.0,"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":0.9994999766349792,"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":0.998199999332428,"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/encoder","display_name":"Encoder","score":0.8780483603477478},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7166376709938049},{"id":"https://openalex.org/keywords/distillation","display_name":"Distillation","score":0.6998619437217712},{"id":"https://openalex.org/keywords/transducer","display_name":"Transducer","score":0.6938676834106445},{"id":"https://openalex.org/keywords/word","display_name":"Word (group theory)","score":0.469011515378952},{"id":"https://openalex.org/keywords/rotary-encoder","display_name":"Rotary encoder","score":0.45806530117988586},{"id":"https://openalex.org/keywords/entropy","display_name":"Entropy (arrow of time)","score":0.4452388286590576},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.4449399411678314},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.37303614616394043},{"id":"https://openalex.org/keywords/acoustics","display_name":"Acoustics","score":0.18472254276275635},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.15496817231178284},{"id":"https://openalex.org/keywords/physics","display_name":"Physics","score":0.07092252373695374}],"concepts":[{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.8780483603477478},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7166376709938049},{"id":"https://openalex.org/C204030448","wikidata":"https://www.wikidata.org/wiki/Q101017","display_name":"Distillation","level":2,"score":0.6998619437217712},{"id":"https://openalex.org/C56318395","wikidata":"https://www.wikidata.org/wiki/Q215928","display_name":"Transducer","level":2,"score":0.6938676834106445},{"id":"https://openalex.org/C90805587","wikidata":"https://www.wikidata.org/wiki/Q10944557","display_name":"Word (group theory)","level":2,"score":0.469011515378952},{"id":"https://openalex.org/C37488316","wikidata":"https://www.wikidata.org/wiki/Q53699","display_name":"Rotary encoder","level":3,"score":0.45806530117988586},{"id":"https://openalex.org/C106301342","wikidata":"https://www.wikidata.org/wiki/Q4117933","display_name":"Entropy (arrow of time)","level":2,"score":0.4452388286590576},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.4449399411678314},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.37303614616394043},{"id":"https://openalex.org/C24890656","wikidata":"https://www.wikidata.org/wiki/Q82811","display_name":"Acoustics","level":1,"score":0.18472254276275635},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.15496817231178284},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.07092252373695374},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0},{"id":"https://openalex.org/C178790620","wikidata":"https://www.wikidata.org/wiki/Q11351","display_name":"Organic chemistry","level":1,"score":0.0},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","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/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.21437/interspeech.2021-797","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2021-797","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2021","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2106.07734","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2106.07734","pdf_url":"https://arxiv.org/pdf/2106.07734","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":"public-domain","license_id":"https://openalex.org/licenses/public-domain","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"mag:3170968215","is_oa":true,"landing_page_url":"https://arxiv.org/pdf/2106.07734.pdf","pdf_url":null,"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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"arXiv (Cornell University)","raw_type":null},{"id":"doi:10.48550/arxiv.2106.07734","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2106.07734","pdf_url":null,"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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2106.07734","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2106.07734","pdf_url":"https://arxiv.org/pdf/2106.07734","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":"public-domain","license_id":"https://openalex.org/licenses/public-domain","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","score":0.8399999737739563,"display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3170968215.pdf","grobid_xml":"https://content.openalex.org/works/W3170968215.grobid-xml"},"referenced_works_count":32,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W1689711448","https://openalex.org/W1821462560","https://openalex.org/W1828163288","https://openalex.org/W2064675550","https://openalex.org/W2127141656","https://openalex.org/W2294543795","https://openalex.org/W2312434537","https://openalex.org/W2508418541","https://openalex.org/W2750499125","https://openalex.org/W2769312802","https://openalex.org/W2794209590","https://openalex.org/W2962760690","https://openalex.org/W2962907457","https://openalex.org/W2963174729","https://openalex.org/W2963376890","https://openalex.org/W2963813662","https://openalex.org/W2973040747","https://openalex.org/W2975044525","https://openalex.org/W2981698279","https://openalex.org/W2990231018","https://openalex.org/W3007227084","https://openalex.org/W3015194534","https://openalex.org/W3096518646","https://openalex.org/W3097522836","https://openalex.org/W3128478537","https://openalex.org/W3144557079","https://openalex.org/W3151287998","https://openalex.org/W3162309234","https://openalex.org/W3162649911","https://openalex.org/W3163062345","https://openalex.org/W3167178529"],"related_works":["https://openalex.org/W3142529478","https://openalex.org/W3038868461","https://openalex.org/W3197201410","https://openalex.org/W3008370744","https://openalex.org/W2607361225","https://openalex.org/W2946555236","https://openalex.org/W2894651928","https://openalex.org/W2992632249","https://openalex.org/W2887264325","https://openalex.org/W2983853253","https://openalex.org/W3180155749","https://openalex.org/W3129868806","https://openalex.org/W2920831951","https://openalex.org/W3094929291","https://openalex.org/W2740284003","https://openalex.org/W2883586237","https://openalex.org/W2922298118","https://openalex.org/W3163983014","https://openalex.org/W3095773170","https://openalex.org/W3141464856"],"abstract_inverted_index":{"We":[0,18,84,109],"propose":[1],"a":[2,27,70,103],"simple":[3],"yet":[4],"effective":[5],"method":[6],"to":[7,52,64,117],"compress":[8],"an":[9,61,95,112],"RNN-Transducer":[10],"(RNN-T)":[11],"through":[12],"the":[13,21,47,54,66,100,124,130],"well-known":[14],"knowledge":[15],"distillation":[16,81,98],"paradigm.":[17],"show":[19,135],"that":[20,86,121],"transducer's":[22,72],"encoder":[23,67,80,97],"outputs":[24,51],"naturally":[25],"have":[26],"high":[28],"entropy":[29,49],"and":[30,74,91,105,126,147],"contain":[31],"rich":[32,40],"information":[33,41],"about":[34],"acoustically":[35],"similar":[36],"word-piece":[37],"confusions.":[38],"This":[39],"is":[42],"suppressed":[43],"when":[44,123],"combined":[45],"with":[46,94],"lower":[48],"decoder":[50],"produce":[53],"joint":[55],"network":[56],"logits.":[57],"Consequently,":[58],"we":[59,115],"introduce":[60],"auxiliary":[62],"loss":[63],"distill":[65],"logits":[68],"from":[69],"teacher":[71,90,107,125],"encoder,":[73],"explore":[75],"training":[76,88],"strategies":[77],"where":[78],"this":[79],"works":[82],"effectively.":[83],"find":[85],"tandem":[87],"of":[89,102],"student":[92,127],"encoders":[93,128],"inplace":[96],"outperforms":[99],"use":[101],"pre-trained":[104],"static":[106],"transducer.":[108],"also":[110],"report":[111],"interesting":[113],"phenomenon":[114],"refer":[116],"as":[118],"implicit":[119],"distillation,":[120],"occurs":[122],"share":[129],"same":[131],"decoder.":[132],"Our":[133],"experiments":[134],"5.37-8.4%":[136],"relative":[137,149],"word":[138],"error":[139],"rate":[140],"reductions":[141],"(WERR)":[142],"on":[143,151],"in-house":[144],"test":[145,153],"sets,":[146],"5.05-6.18%":[148],"WERRs":[150],"LibriSpeech":[152],"sets.":[154]},"counts_by_year":[{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":1},{"year":2022,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}