{"id":"https://openalex.org/W3197976839","doi":"https://doi.org/10.21437/interspeech.2021-212","title":"Tied & Reduced RNN-T Decoder","display_name":"Tied & Reduced RNN-T Decoder","publication_year":2021,"publication_date":"2021-08-27","ids":{"openalex":"https://openalex.org/W3197976839","doi":"https://doi.org/10.21437/interspeech.2021-212","mag":"3197976839"},"language":"en","primary_location":{"id":"doi:10.21437/interspeech.2021-212","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2021-212","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":"article","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2109.07513","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5008863473","display_name":"Rami Botros","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Rami Botros","raw_affiliation_strings":["Google Inc., U"],"affiliations":[{"raw_affiliation_string":"Google Inc., U","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5070513394","display_name":"Tara N. Sainath","orcid":"https://orcid.org/0000-0002-4126-6556"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Tara N. Sainath","raw_affiliation_strings":["Google Inc., U"],"affiliations":[{"raw_affiliation_string":"Google Inc., U","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5062218486","display_name":"Robert David","orcid":"https://orcid.org/0000-0002-3244-5341"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Robert David","raw_affiliation_strings":["Google Inc., U"],"affiliations":[{"raw_affiliation_string":"Google Inc., U","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5020536207","display_name":"Emmanuel Guzman","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Emmanuel Guzman","raw_affiliation_strings":["Google Inc., U"],"affiliations":[{"raw_affiliation_string":"Google Inc., U","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100318020","display_name":"Wei Li","orcid":"https://orcid.org/0000-0001-7824-4839"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Wei Li","raw_affiliation_strings":["Google Inc., U"],"affiliations":[{"raw_affiliation_string":"Google Inc., U","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101319167","display_name":"Yanzhang He","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yanzhang He","raw_affiliation_strings":["Google Inc., U"],"affiliations":[{"raw_affiliation_string":"Google Inc., U","institution_ids":["https://openalex.org/I1291425158"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5008863473"],"corresponding_institution_ids":["https://openalex.org/I1291425158"],"apc_list":null,"apc_paid":null,"fwci":3.5203,"has_fulltext":false,"cited_by_count":27,"citation_normalized_percentile":{"value":0.93829303,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":95,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"4563","last_page":"4567"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9998000264167786,"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":0.9998000264167786,"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":0.9998000264167786,"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":0.9991000294685364,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/recurrent-neural-network","display_name":"Recurrent neural network","score":0.7964372038841248},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7871648073196411},{"id":"https://openalex.org/keywords/decoding-methods","display_name":"Decoding methods","score":0.5960410237312317},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.5840469002723694},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5777245163917542},{"id":"https://openalex.org/keywords/word","display_name":"Word (group theory)","score":0.504116415977478},{"id":"https://openalex.org/keywords/reduction","display_name":"Reduction (mathematics)","score":0.4878561496734619},{"id":"https://openalex.org/keywords/embedding","display_name":"Embedding","score":0.4706895649433136},{"id":"https://openalex.org/keywords/dropout","display_name":"Dropout (neural networks)","score":0.46353745460510254},{"id":"https://openalex.org/keywords/bayes-theorem","display_name":"Bayes' theorem","score":0.45086273550987244},{"id":"https://openalex.org/keywords/word-error-rate","display_name":"Word error rate","score":0.4354839324951172},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.4311710596084595},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.39041978120803833},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.35220658779144287},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.34577375650405884},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.24028873443603516},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.11422690749168396}],"concepts":[{"id":"https://openalex.org/C147168706","wikidata":"https://www.wikidata.org/wiki/Q1457734","display_name":"Recurrent neural network","level":3,"score":0.7964372038841248},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7871648073196411},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.5960410237312317},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.5840469002723694},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5777245163917542},{"id":"https://openalex.org/C90805587","wikidata":"https://www.wikidata.org/wiki/Q10944557","display_name":"Word (group theory)","level":2,"score":0.504116415977478},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.4878561496734619},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.4706895649433136},{"id":"https://openalex.org/C2776145597","wikidata":"https://www.wikidata.org/wiki/Q25339462","display_name":"Dropout (neural networks)","level":2,"score":0.46353745460510254},{"id":"https://openalex.org/C207201462","wikidata":"https://www.wikidata.org/wiki/Q182505","display_name":"Bayes' theorem","level":3,"score":0.45086273550987244},{"id":"https://openalex.org/C40969351","wikidata":"https://www.wikidata.org/wiki/Q3516228","display_name":"Word error rate","level":2,"score":0.4354839324951172},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.4311710596084595},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.39041978120803833},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.35220658779144287},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.34577375650405884},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.24028873443603516},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.11422690749168396},{"id":"https://openalex.org/C107673813","wikidata":"https://www.wikidata.org/wiki/Q812534","display_name":"Bayesian probability","level":2,"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/C151730666","wikidata":"https://www.wikidata.org/wiki/Q7205","display_name":"Paleontology","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}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.21437/interspeech.2021-212","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2021-212","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:2109.07513","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2109.07513","pdf_url":"https://arxiv.org/pdf/2109.07513","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":null,"raw_type":"text"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2109.07513","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2109.07513","pdf_url":"https://arxiv.org/pdf/2109.07513","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":null,"raw_type":"text"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":43,"referenced_works":["https://openalex.org/W950853366","https://openalex.org/W1828163288","https://openalex.org/W1855892484","https://openalex.org/W2013598660","https://openalex.org/W2033256038","https://openalex.org/W2515439472","https://openalex.org/W2545177271","https://openalex.org/W2549416390","https://openalex.org/W2617258110","https://openalex.org/W2626778328","https://openalex.org/W2747467128","https://openalex.org/W2767286248","https://openalex.org/W2928941594","https://openalex.org/W2936774411","https://openalex.org/W2951327905","https://openalex.org/W2962760690","https://openalex.org/W2962824709","https://openalex.org/W2962826786","https://openalex.org/W2963747784","https://openalex.org/W2990501692","https://openalex.org/W3007227084","https://openalex.org/W3008181812","https://openalex.org/W3015194534","https://openalex.org/W3015686596","https://openalex.org/W3015995734","https://openalex.org/W3016010032","https://openalex.org/W3016234571","https://openalex.org/W3025165719","https://openalex.org/W3032859754","https://openalex.org/W3094979069","https://openalex.org/W3095311338","https://openalex.org/W3095697114","https://openalex.org/W3097777922","https://openalex.org/W3111191378","https://openalex.org/W3124059949","https://openalex.org/W3132164338","https://openalex.org/W3160405885","https://openalex.org/W3160766462","https://openalex.org/W3161375121","https://openalex.org/W3161532853","https://openalex.org/W4320930577","https://openalex.org/W4385245566","https://openalex.org/W4394666973"],"related_works":["https://openalex.org/W3082178636","https://openalex.org/W2782041652","https://openalex.org/W2612657834","https://openalex.org/W4298287631","https://openalex.org/W2953061907","https://openalex.org/W2392157706","https://openalex.org/W2599192953","https://openalex.org/W2952088488","https://openalex.org/W1973600295","https://openalex.org/W1599512561"],"abstract_inverted_index":{"Previous":[0],"works":[1],"on":[2],"the":[3,41,88,113,123,154],"Recurrent":[4],"Neural":[5],"Network-Transducer":[6],"(RNN-T)":[7],"models":[8],"have":[9],"shown":[10],"that,":[11],"under":[12],"some":[13],"conditions,":[14],"it":[15],"is":[16,37],"possible":[17],"to":[18,86,160],"simplify":[19],"its":[20,53,118],"prediction":[21,105],"network":[22,92,106],"with":[23,122,145],"little":[24],"or":[25],"no":[26],"loss":[27],"in":[28,65,101,133,143],"recognition":[29,102],"accuracy":[30],"(arXiv:2003.07705":[31],"[eess.AS],":[32],"[2],":[33],"arXiv:2012.06749":[34],"[cs.CL]).":[35],"This":[36,138],"done":[38],"by":[39],"limiting":[40],"context":[42],"size":[43],"of":[44,56,60,112],"previous":[45],"labels":[46],"and/or":[47],"using":[48],"a":[49,108],"simpler":[50],"architecture":[51],"for":[52,77],"layers":[54],"instead":[55],"LSTMs.":[57],"The":[58],"benefits":[59],"such":[61],"changes":[62],"include":[63],"reduction":[64],"model":[66],"size,":[67],"faster":[68,98],"inference":[69],"and":[70,97,116],"power":[71],"savings,":[72],"which":[73],"are":[74],"all":[75],"useful":[76],"on-device":[78],"applications.":[79],"In":[80],"this":[81],"work,":[82],"we":[83],"study":[84],"ways":[85],"make":[87],"RNN-T":[89,155],"decoder":[90],"(prediction":[91],"+":[93],"joint":[94,124],"network)":[95],"smaller":[96],"without":[99,163],"degradation":[100],"performance.":[103],"Our":[104],"performs":[107],"simple":[109,139],"weighted":[110],"averaging":[111],"input":[114],"embeddings,":[115],"shares":[117],"embedding":[119],"matrix":[120],"weights":[121],"network's":[125],"output":[126],"layer":[127],"(a.k.a.":[128],"weight":[129],"tying,":[130],"commonly":[131],"used":[132,142],"language":[134],"modeling":[135],"arXiv:1611.01462":[136],"[cs.LG]).":[137],"design,":[140],"when":[141],"conjunction":[144],"additional":[146],"Edit-based":[147],"Minimum":[148],"Bayes":[149],"Risk":[150],"(EMBR)":[151],"training,":[152],"reduces":[153],"Decoder":[156],"from":[157],"23M":[158],"parameters":[159],"just":[161],"2M,":[162],"affecting":[164],"word-error":[165],"rate":[166],"(WER).":[167]},"counts_by_year":[{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":4},{"year":2023,"cited_by_count":18},{"year":2022,"cited_by_count":3}],"updated_date":"2026-04-03T22:45:19.894376","created_date":"2021-09-13T00:00:00"}