{"id":"https://openalex.org/W3005302685","doi":"https://doi.org/10.1109/lsp.2020.3036335","title":"Accelerating RNN Transducer Inference via Adaptive Expansion Search","display_name":"Accelerating RNN Transducer Inference via Adaptive Expansion Search","publication_year":2020,"publication_date":"2020-01-01","ids":{"openalex":"https://openalex.org/W3005302685","doi":"https://doi.org/10.1109/lsp.2020.3036335","mag":"3005302685"},"language":"en","primary_location":{"id":"doi:10.1109/lsp.2020.3036335","is_oa":false,"landing_page_url":"https://doi.org/10.1109/lsp.2020.3036335","pdf_url":null,"source":{"id":"https://openalex.org/S120629676","display_name":"IEEE Signal Processing Letters","issn_l":"1070-9908","issn":["1070-9908","1558-2361"],"is_oa":false,"is_in_doaj":false,"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":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Signal Processing Letters","raw_type":"journal-article"},"type":"article","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2002.03577","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Juntae Kim","orcid":"https://orcid.org/0000-0003-3344-4591"},"institutions":[{"id":"https://openalex.org/I4210114081","display_name":"Korea Kacoh (South Korea)","ror":"https://ror.org/02930yz47","country_code":"KR","type":"company","lineage":["https://openalex.org/I4210114081"]}],"countries":["KR"],"is_corresponding":true,"raw_author_name":"Juntae Kim","raw_affiliation_strings":["Kakao Enterprise and Kakao Corporation, Gyeonggi-do, South Korea"],"affiliations":[{"raw_affiliation_string":"Kakao Enterprise and Kakao Corporation, Gyeonggi-do, South Korea","institution_ids":["https://openalex.org/I4210114081"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Yoonhan Lee","orcid":null},"institutions":[{"id":"https://openalex.org/I4210114081","display_name":"Korea Kacoh (South Korea)","ror":"https://ror.org/02930yz47","country_code":"KR","type":"company","lineage":["https://openalex.org/I4210114081"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Yoonhan Lee","raw_affiliation_strings":["Kakao Enterprise and Kakao Corporation, Gyeonggi-do, South Korea"],"affiliations":[{"raw_affiliation_string":"Kakao Enterprise and Kakao Corporation, Gyeonggi-do, South Korea","institution_ids":["https://openalex.org/I4210114081"]}]},{"author_position":"last","author":{"id":null,"display_name":"Eesung Kim","orcid":null},"institutions":[{"id":"https://openalex.org/I4210114081","display_name":"Korea Kacoh (South Korea)","ror":"https://ror.org/02930yz47","country_code":"KR","type":"company","lineage":["https://openalex.org/I4210114081"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Eesung Kim","raw_affiliation_strings":["Kakao Enterprise and Kakao Corporation, Gyeonggi-do, South Korea"],"affiliations":[{"raw_affiliation_string":"Kakao Enterprise and Kakao Corporation, Gyeonggi-do, South Korea","institution_ids":["https://openalex.org/I4210114081"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I4210114081"],"apc_list":null,"apc_paid":null,"fwci":1.6456,"has_fulltext":false,"cited_by_count":21,"citation_normalized_percentile":{"value":0.8696204,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":98},"biblio":{"volume":"27","issue":null,"first_page":"2019","last_page":"2023"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.41769999265670776,"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":0.41769999265670776,"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/T10320","display_name":"Neural Networks and Applications","score":0.1428000032901764,"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":0.1371999979019165,"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/decoding-methods","display_name":"Decoding methods","score":0.7989000082015991},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.6735000014305115},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.5819000005722046},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.5235999822616577},{"id":"https://openalex.org/keywords/recurrent-neural-network","display_name":"Recurrent neural network","score":0.4909999966621399},{"id":"https://openalex.org/keywords/sequence","display_name":"Sequence (biology)","score":0.48559999465942383},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.4812999963760376},{"id":"https://openalex.org/keywords/word","display_name":"Word (group theory)","score":0.4659999907016754},{"id":"https://openalex.org/keywords/character","display_name":"Character (mathematics)","score":0.43309998512268066}],"concepts":[{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.7989000082015991},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7093999981880188},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.6735000014305115},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.5819000005722046},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.5235999822616577},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4934000074863434},{"id":"https://openalex.org/C147168706","wikidata":"https://www.wikidata.org/wiki/Q1457734","display_name":"Recurrent neural network","level":3,"score":0.4909999966621399},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.48559999465942383},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.4812999963760376},{"id":"https://openalex.org/C90805587","wikidata":"https://www.wikidata.org/wiki/Q10944557","display_name":"Word (group theory)","level":2,"score":0.4659999907016754},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.4334999918937683},{"id":"https://openalex.org/C2780861071","wikidata":"https://www.wikidata.org/wiki/Q1062934","display_name":"Character (mathematics)","level":2,"score":0.43309998512268066},{"id":"https://openalex.org/C125583679","wikidata":"https://www.wikidata.org/wiki/Q755673","display_name":"Search algorithm","level":2,"score":0.382099986076355},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3684000074863434},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.349700003862381},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.3407000005245209},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.3370000123977661},{"id":"https://openalex.org/C19889080","wikidata":"https://www.wikidata.org/wiki/Q2835852","display_name":"Beam search","level":3,"score":0.3203999996185303},{"id":"https://openalex.org/C204259536","wikidata":"https://www.wikidata.org/wiki/Q787903","display_name":"Linear search","level":2,"score":0.3183000087738037},{"id":"https://openalex.org/C831591","wikidata":"https://www.wikidata.org/wiki/Q59750","display_name":"Bidirectional search","level":5,"score":0.3100999891757965},{"id":"https://openalex.org/C193969084","wikidata":"https://www.wikidata.org/wiki/Q7452500","display_name":"Sequential decoding","level":4,"score":0.3005000054836273},{"id":"https://openalex.org/C204397858","wikidata":"https://www.wikidata.org/wiki/Q4437907","display_name":"List decoding","level":5,"score":0.28769999742507935},{"id":"https://openalex.org/C40743351","wikidata":"https://www.wikidata.org/wiki/Q7002049","display_name":"Neural decoding","level":3,"score":0.2775000035762787},{"id":"https://openalex.org/C40969351","wikidata":"https://www.wikidata.org/wiki/Q3516228","display_name":"Word error rate","level":2,"score":0.25929999351501465},{"id":"https://openalex.org/C143724316","wikidata":"https://www.wikidata.org/wiki/Q312468","display_name":"Series (stratigraphy)","level":2,"score":0.25600001215934753}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/lsp.2020.3036335","is_oa":false,"landing_page_url":"https://doi.org/10.1109/lsp.2020.3036335","pdf_url":null,"source":{"id":"https://openalex.org/S120629676","display_name":"IEEE Signal Processing Letters","issn_l":"1070-9908","issn":["1070-9908","1558-2361"],"is_oa":false,"is_in_doaj":false,"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":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Signal Processing Letters","raw_type":"journal-article"},{"id":"pmh:oai:arXiv.org:2002.03577","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2002.03577","pdf_url":"https://arxiv.org/pdf/2002.03577","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:2002.03577","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2002.03577","pdf_url":"https://arxiv.org/pdf/2002.03577","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":18,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W1967325250","https://openalex.org/W2160815625","https://openalex.org/W2327501763","https://openalex.org/W2519224033","https://openalex.org/W2746192915","https://openalex.org/W2962760690","https://openalex.org/W2962780374","https://openalex.org/W2962824709","https://openalex.org/W2963070863","https://openalex.org/W2963250244","https://openalex.org/W2963414781","https://openalex.org/W2963970535","https://openalex.org/W3008174054","https://openalex.org/W6638749077","https://openalex.org/W6756419475","https://openalex.org/W6769806307","https://openalex.org/W6770245836"],"related_works":[],"abstract_inverted_index":{"Recurrent":[0],"neural":[1],"network":[2],"transducers":[3],"(RNN-T)":[4],"are":[5],"a":[6,18,37,61,82,131,163],"promising":[7],"end-to-end":[8],"speech":[9],"recognition":[10],"framework":[11],"that":[12,96,134],"transduce":[13],"input":[14],"acoustic":[15],"frames":[16,79],"to":[17,81,116,137,153],"character":[19,90],"sequence.":[20],"Best-":[21],"and":[22,129,162],"breadth-first":[23,51],"searches":[24],"have":[25],"been":[26],"used":[27],"as":[28],"decoding":[29,48,72,78,99],"strategies":[30],"for":[31,40,70,155],"RNN-T.":[32],"However,":[33],"best-first":[34,58],"search":[35,52,59,69,114],"follows":[36],"sequential":[38,55],"process":[39,56,133],"its":[41],"expansion":[42,68,113,140,147],"search,":[43],"which":[44],"slows":[45],"down":[46],"the":[47,54,77,86,89,98,122,127,139],"process.":[49],"Although":[50],"replaces":[53],"of":[57,76,88,97],"with":[60,169],"parallel":[62],"one,":[63],"it":[64],"unnecessarily":[65],"conducts":[66],"an":[67,111],"all":[71],"steps.":[73],"As":[74],"most":[75],"correspond":[80],"blank":[83],"symbol":[84],"because":[85],"length":[87],"sequence":[91],"is":[92,151],"much":[93],"shorter":[94],"than":[95],"frames,":[100],"this":[101],"induces":[102],"computational":[103],"overhead.":[104],"To":[105],"address":[106],"these":[107],"limitations,":[108],"we":[109],"introduce":[110],"adaptive":[112],"(AES)":[115],"accelerate":[117],"RNN-T":[118],"inference.":[119],"AES":[120,143,154],"overcomes":[121],"aforementioned":[123],"limitations":[124],"by":[125],"batching":[126],"hypotheses":[128],"adopting":[130],"decision-making":[132],"decides":[135],"whether":[136],"continue":[138],"search;":[141],"thus,":[142],"can":[144],"avoid":[145],"unnecessary":[146],"search.":[148],"Furthermore,":[149],"pruning":[150],"applied":[152],"further":[156],"acceleration.":[157],"We":[158],"achieved":[159],"significant":[160],"speedup":[161],"lower":[164],"word":[165],"error":[166],"rate":[167],"compared":[168],"other":[170],"baselines.":[171]},"counts_by_year":[{"year":2025,"cited_by_count":5},{"year":2024,"cited_by_count":4},{"year":2023,"cited_by_count":5},{"year":2022,"cited_by_count":2},{"year":2021,"cited_by_count":5}],"updated_date":"2026-03-20T23:20:44.827607","created_date":"2020-02-14T00:00:00"}