{"id":"https://openalex.org/W4392902568","doi":"https://doi.org/10.1109/icassp48485.2024.10446861","title":"Stateful Conformer with Cache-Based Inference for Streaming Automatic Speech Recognition","display_name":"Stateful Conformer with Cache-Based Inference for Streaming Automatic Speech Recognition","publication_year":2024,"publication_date":"2024-03-18","ids":{"openalex":"https://openalex.org/W4392902568","doi":"https://doi.org/10.1109/icassp48485.2024.10446861"},"language":"en","primary_location":{"id":"doi:10.1109/icassp48485.2024.10446861","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp48485.2024.10446861","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2024 - 2024 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/A5005843015","display_name":"Vahid Noroozi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Vahid Noroozi","raw_affiliation_strings":["NVIDIA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NVIDIA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5027823948","display_name":"Somshubra Majumdar","orcid":"https://orcid.org/0000-0001-5635-4893"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Somshubra Majumdar","raw_affiliation_strings":["NVIDIA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NVIDIA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100699893","display_name":"Ankur Kumar","orcid":"https://orcid.org/0000-0002-2417-9581"},"institutions":[{"id":"https://openalex.org/I4210156583","display_name":"Laboratoire d'Informatique de Paris-Nord","ror":"https://ror.org/05g1zjw44","country_code":"FR","type":"facility","lineage":["https://openalex.org/I1294671590","https://openalex.org/I1294671590","https://openalex.org/I4210091279","https://openalex.org/I4210156583","https://openalex.org/I4210159245"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Ankur Kumar","raw_affiliation_strings":["UCLA,Department of Computer Science","Department of Computer Science, UCLA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UCLA,Department of Computer Science","institution_ids":["https://openalex.org/I4210156583"]},{"raw_affiliation_string":"Department of Computer Science, UCLA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5040747392","display_name":"Jagadeesh Balam","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jagadeesh Balam","raw_affiliation_strings":["NVIDIA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NVIDIA","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5032957280","display_name":"Boris Ginsburg","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Boris Ginsburg","raw_affiliation_strings":["NVIDIA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NVIDIA","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":11,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"12041","last_page":"12045"},"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":0.9994999766349792,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8781486749649048},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.7876818180084229},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.6036868095397949},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.5452549457550049},{"id":"https://openalex.org/keywords/cache","display_name":"Cache","score":0.5321865081787109},{"id":"https://openalex.org/keywords/autoregressive-model","display_name":"Autoregressive model","score":0.5142301917076111},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.49178436398506165},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4418635070323944},{"id":"https://openalex.org/keywords/stateful-firewall","display_name":"Stateful firewall","score":0.43870043754577637},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.4203597903251648},{"id":"https://openalex.org/keywords/real-time-computing","display_name":"Real-time computing","score":0.32384562492370605},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.1537756323814392}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8781486749649048},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7876818180084229},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.6036868095397949},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.5452549457550049},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.5321865081787109},{"id":"https://openalex.org/C159877910","wikidata":"https://www.wikidata.org/wiki/Q2202883","display_name":"Autoregressive model","level":2,"score":0.5142301917076111},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.49178436398506165},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4418635070323944},{"id":"https://openalex.org/C22927095","wikidata":"https://www.wikidata.org/wiki/Q1784206","display_name":"Stateful firewall","level":3,"score":0.43870043754577637},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.4203597903251648},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.32384562492370605},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.1537756323814392},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.0},{"id":"https://openalex.org/C149782125","wikidata":"https://www.wikidata.org/wiki/Q160039","display_name":"Econometrics","level":1,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"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/C158379750","wikidata":"https://www.wikidata.org/wiki/Q214111","display_name":"Network packet","level":2,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp48485.2024.10446861","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp48485.2024.10446861","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":22,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W2962760690","https://openalex.org/W2963250244","https://openalex.org/W2974231335","https://openalex.org/W3015671919","https://openalex.org/W3015974384","https://openalex.org/W3016010032","https://openalex.org/W3096702180","https://openalex.org/W3096888553","https://openalex.org/W3097125541","https://openalex.org/W3097777922","https://openalex.org/W3148654612","https://openalex.org/W3160766462","https://openalex.org/W3161873870","https://openalex.org/W3162665866","https://openalex.org/W3163907627","https://openalex.org/W3197478142","https://openalex.org/W3198492054","https://openalex.org/W4391021542","https://openalex.org/W6739901393","https://openalex.org/W6767671539","https://openalex.org/W6790121257"],"related_works":["https://openalex.org/W2583381754","https://openalex.org/W2144255176","https://openalex.org/W2222333653","https://openalex.org/W109977192","https://openalex.org/W1970797462","https://openalex.org/W2019740326","https://openalex.org/W2047113772","https://openalex.org/W2566747981","https://openalex.org/W4312120139","https://openalex.org/W2809541058"],"abstract_inverted_index":{"In":[0],"this":[1],"paper,":[2],"we":[3],"propose":[4],"an":[5,41],"efficient":[6],"and":[7,32,38,72,96,105,116],"accurate":[8],"streaming":[9,24,80,124],"speech":[10],"recognition":[11],"model":[12,57,104,125],"based":[13],"on":[14],"the":[15,20,30,36,47,66,70,102],"FastConformer":[16,21],"architecture.":[17],"We":[18,100],"adapted":[19],"architecture":[22],"for":[23,78],"applications":[25],"through:":[26],"(1)":[27],"constraining":[28],"both":[29],"look-ahead":[31],"past":[33],"contexts":[34],"in":[35,61],"encoder,":[37],"(2)":[39],"introducing":[40],"activation":[42],"caching":[43],"mechanism":[44],"to":[45,50,64,120],"enable":[46],"non-autoregressive":[48],"encoder":[49,85],"operate":[51],"autoregressively":[52],"during":[53],"inference.":[54],"The":[55],"proposed":[56,84,103],"is":[58,76],"thoughtfully":[59],"designed":[60],"a":[62,121],"way":[63],"eliminate":[65],"accuracy":[67,112],"disparity":[68],"between":[69],"train":[71],"inference":[73,117],"time":[74,118],"which":[75],"common":[77],"many":[79],"models.":[81],"Furthermore,":[82],"our":[83],"works":[86],"with":[87,113],"various":[88],"decoder":[89],"configurations":[90],"including":[91],"Connectionist":[92],"Temporal":[93],"Classification":[94],"(CTC)":[95],"RNN-Transducer":[97],"(RNNT)":[98],"decoders.":[99],"evaluate":[101],"demonstrate":[106],"that":[107],"it":[108],"can":[109],"achieve":[110],"better":[111],"lower":[114],"latency":[115],"compared":[119],"conventional":[122],"buffered":[123],"baseline.":[126]},"counts_by_year":[{"year":2026,"cited_by_count":3},{"year":2025,"cited_by_count":7},{"year":2024,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}