{"id":"https://openalex.org/W4392904789","doi":"https://doi.org/10.1109/icassp48485.2024.10446584","title":"Improving Speed/Accuracy Tradeoff for Online Streaming ASR via Real-Valued and Trainable Strides","display_name":"Improving Speed/Accuracy Tradeoff for Online Streaming ASR via Real-Valued and Trainable Strides","publication_year":2024,"publication_date":"2024-03-18","ids":{"openalex":"https://openalex.org/W4392904789","doi":"https://doi.org/10.1109/icassp48485.2024.10446584"},"language":"en","primary_location":{"id":"doi:10.1109/icassp48485.2024.10446584","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp48485.2024.10446584","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/A5014480570","display_name":"D. Albesano","orcid":null},"institutions":[{"id":"https://openalex.org/I4210164937","display_name":"Microsoft Research (United Kingdom)","ror":"https://ror.org/05k87vq12","country_code":"GB","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Dario Albesano","raw_affiliation_strings":["Microsoft"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft","institution_ids":["https://openalex.org/I4210164937"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5027968457","display_name":"Nicola Ferri","orcid":"https://orcid.org/0000-0001-8898-7441"},"institutions":[{"id":"https://openalex.org/I4210164937","display_name":"Microsoft Research (United Kingdom)","ror":"https://ror.org/05k87vq12","country_code":"GB","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Nicola Ferri","raw_affiliation_strings":["Microsoft"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft","institution_ids":["https://openalex.org/I4210164937"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5016747567","display_name":"Felix Weninger","orcid":null},"institutions":[{"id":"https://openalex.org/I4210164937","display_name":"Microsoft Research (United Kingdom)","ror":"https://ror.org/05k87vq12","country_code":"GB","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Felix Weninger","raw_affiliation_strings":["Microsoft"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft","institution_ids":["https://openalex.org/I4210164937"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5109588257","display_name":"Puming Zhan","orcid":null},"institutions":[{"id":"https://openalex.org/I4210164937","display_name":"Microsoft Research (United Kingdom)","ror":"https://ror.org/05k87vq12","country_code":"GB","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Puming Zhan","raw_affiliation_strings":["Microsoft"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft","institution_ids":["https://openalex.org/I4210164937"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I4210164937"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"11716","last_page":"11720"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9969000220298767,"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":0.9969000220298767,"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/T11447","display_name":"Blind Source Separation Techniques","score":0.9912999868392944,"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/T10860","display_name":"Speech and Audio Processing","score":0.9909999966621399,"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/upsampling","display_name":"Upsampling","score":0.8512028455734253},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7884384393692017},{"id":"https://openalex.org/keywords/stride","display_name":"STRIDE","score":0.6613718271255493},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5557027459144592},{"id":"https://openalex.org/keywords/pooling","display_name":"Pooling","score":0.5527957677841187},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.5506939888000488},{"id":"https://openalex.org/keywords/discretization","display_name":"Discretization","score":0.4357500374317169},{"id":"https://openalex.org/keywords/stochastic-game","display_name":"Stochastic game","score":0.420012891292572},{"id":"https://openalex.org/keywords/flexibility","display_name":"Flexibility (engineering)","score":0.41823506355285645},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.41384586691856384},{"id":"https://openalex.org/keywords/convolution","display_name":"Convolution (computer science)","score":0.4104387164115906},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.41005682945251465},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.39098620414733887},{"id":"https://openalex.org/keywords/real-time-computing","display_name":"Real-time computing","score":0.37385350465774536},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.3298880457878113},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.16286858916282654},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.13077762722969055},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.1271313726902008}],"concepts":[{"id":"https://openalex.org/C110384440","wikidata":"https://www.wikidata.org/wiki/Q1143270","display_name":"Upsampling","level":3,"score":0.8512028455734253},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7884384393692017},{"id":"https://openalex.org/C18007350","wikidata":"https://www.wikidata.org/wiki/Q7394815","display_name":"STRIDE","level":2,"score":0.6613718271255493},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5557027459144592},{"id":"https://openalex.org/C70437156","wikidata":"https://www.wikidata.org/wiki/Q7228652","display_name":"Pooling","level":2,"score":0.5527957677841187},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.5506939888000488},{"id":"https://openalex.org/C73000952","wikidata":"https://www.wikidata.org/wiki/Q17007827","display_name":"Discretization","level":2,"score":0.4357500374317169},{"id":"https://openalex.org/C22171661","wikidata":"https://www.wikidata.org/wiki/Q1074380","display_name":"Stochastic game","level":2,"score":0.420012891292572},{"id":"https://openalex.org/C2780598303","wikidata":"https://www.wikidata.org/wiki/Q65921492","display_name":"Flexibility (engineering)","level":2,"score":0.41823506355285645},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.41384586691856384},{"id":"https://openalex.org/C45347329","wikidata":"https://www.wikidata.org/wiki/Q5166604","display_name":"Convolution (computer science)","level":3,"score":0.4104387164115906},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.41005682945251465},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.39098620414733887},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.37385350465774536},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3298880457878113},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.16286858916282654},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.13077762722969055},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.1271313726902008},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.0},{"id":"https://openalex.org/C144237770","wikidata":"https://www.wikidata.org/wiki/Q747534","display_name":"Mathematical economics","level":1,"score":0.0},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","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/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.0},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp48485.2024.10446584","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp48485.2024.10446584","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":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":30,"referenced_works":["https://openalex.org/W2750499125","https://openalex.org/W2914304175","https://openalex.org/W2963827914","https://openalex.org/W3007433671","https://openalex.org/W3015974384","https://openalex.org/W3016010032","https://openalex.org/W3097777922","https://openalex.org/W3123221884","https://openalex.org/W3161873870","https://openalex.org/W3162665866","https://openalex.org/W3197813307","https://openalex.org/W3197917733","https://openalex.org/W3197976839","https://openalex.org/W3211278025","https://openalex.org/W4220727198","https://openalex.org/W4239396298","https://openalex.org/W4281779489","https://openalex.org/W4283741996","https://openalex.org/W4284881636","https://openalex.org/W4288088457","https://openalex.org/W4372259771","https://openalex.org/W4385245566","https://openalex.org/W4391021542","https://openalex.org/W6631190155","https://openalex.org/W6638749077","https://openalex.org/W6638946568","https://openalex.org/W6769806307","https://openalex.org/W6789463137","https://openalex.org/W6793959113","https://openalex.org/W6838276489"],"related_works":["https://openalex.org/W2058965144","https://openalex.org/W2559156603","https://openalex.org/W4300832495","https://openalex.org/W2987852271","https://openalex.org/W4287394948","https://openalex.org/W2967990525","https://openalex.org/W2949066288","https://openalex.org/W3119356360","https://openalex.org/W3202075396","https://openalex.org/W4214604401"],"abstract_inverted_index":{"The":[0,73],"Conformer":[1],"Transducer":[2],"(CT)":[3],"is":[4,40],"arguably":[5],"the":[6,23,29,33,46,52,57,61,91,106,113,122,130,145,149],"most":[7],"popular":[8],"architecture":[9],"for":[10,27,135],"online":[11,116],"streaming":[12],"end-to-end":[13],"(E2E)":[14],"ASR":[15,118,138],"systems.":[16],"Since":[17],"it":[18,89],"has":[19,66],"quadratic":[20],"complexity":[21],"in":[22,56],"input":[24,34],"sequence":[25,35],"length":[26,39],"computing":[28,47],"attention":[30],"weights,":[31],"downsampling":[32,59,81],"to":[36,44,67,112,143,182],"reduce":[37],"its":[38],"an":[41,156],"effective":[42],"way":[43],"mitigate":[45],"cost":[48],"and":[49,109,140,173],"speed":[50,74,97,175],"up":[51,75],"inference":[53,174],"process.":[54],"However,":[55],"traditional":[58],"approach,":[60],"sampling":[62],"factor":[63,193],"(i.e.":[64],"stride)":[65],"be":[68],"a":[69,124,198],"pre-defined":[70],"integer":[71,186],"value.":[72],"achieved":[76],"by":[77,176,194],"such":[78],"kind":[79],"of":[80,93,132],"often":[82],"comes":[83],"with":[84,96,148,185,201],"significant":[85],"accuracy":[86,95,172,207],"degradation,":[87],"because":[88],"lacks":[90],"flexibility":[92],"trading":[94],"at":[98],"fine-grained":[99],"level.":[100],"In":[101],"this":[102],"paper,":[103],"we":[104,165],"apply":[105],"spectral":[107],"pooling":[108],"DiffStride":[110],"techniques":[111,134],"CT":[114,136],"based":[115,137],"E2E":[117],"system.":[119],"This":[120],"makes":[121],"stride":[123,146,179,187],"real-valued":[125,178],"trainable":[126],"parameter.":[127,180],"We":[128,152],"optimize":[129],"implementation":[131],"these":[133],"systems":[139],"develop":[141],"recipes":[142],"train":[144],"together":[147],"model":[150],"parameters.":[151],"conduct":[153],"experiments":[154],"on":[155,197],"internal":[157],"medical":[158,199],"conversation":[159],"dataset.":[160],"Our":[161],"results":[162],"show":[163],"that":[164],"can":[166],"achieve":[167],"better":[168],"tradeoff":[169],"between":[170],"recognition":[171],"training":[177],"Compared":[181],"using":[183],"decimation":[184],"value,":[188],"our":[189],"approach":[190],"reduces":[191],"real-time":[192],"15.6":[195],"%":[196,205],"dataset":[200],"less":[202],"than":[203],"1":[204],"relative":[206],"degradation.":[208]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}