{"id":"https://openalex.org/W4284881636","doi":"https://doi.org/10.21437/interspeech.2022-10465","title":"Compute Cost Amortized Transformer for Streaming ASR","display_name":"Compute Cost Amortized Transformer for Streaming ASR","publication_year":2022,"publication_date":"2022-09-16","ids":{"openalex":"https://openalex.org/W4284881636","doi":"https://doi.org/10.21437/interspeech.2022-10465"},"language":"en","primary_location":{"id":"doi:10.21437/interspeech.2022-10465","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2022-10465","pdf_url":null,"source":{"id":"https://openalex.org/S4363604309","display_name":"Interspeech 2022","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2022","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/A5068891302","display_name":"Yi Xie","orcid":"https://orcid.org/0000-0002-1416-5557"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yi Xie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5082890917","display_name":"Jonathan J. Macoskey","orcid":"https://orcid.org/0000-0002-2776-230X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jonathan J. Macoskey","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5023415990","display_name":"Martin Radfar","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Martin Radfar","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102963746","display_name":"Feng-Ju Chang","orcid":"https://orcid.org/0000-0003-2405-3118"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Feng-Ju Chang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5027102522","display_name":"Brian King","orcid":"https://orcid.org/0000-0002-5300-5564"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Brian King","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5110230355","display_name":"Ariya Rastrow","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ariya Rastrow","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066042040","display_name":"Athanasios Mouchtaris","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Athanasios Mouchtaris","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5023708721","display_name":"Grant P. Strimel","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Grant Strimel","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":6,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"3043","last_page":"3047"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9991999864578247,"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":0.9991999864578247,"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":0.9965000152587891,"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/T10320","display_name":"Neural Networks and Applications","score":0.9922999739646912,"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/computer-science","display_name":"Computer science","score":0.6484482884407043},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.5639519095420837},{"id":"https://openalex.org/keywords/electrical-engineering","display_name":"Electrical engineering","score":0.21080723404884338},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.13914042711257935},{"id":"https://openalex.org/keywords/voltage","display_name":"Voltage","score":0.05257174372673035}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6484482884407043},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.5639519095420837},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.21080723404884338},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.13914042711257935},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.05257174372673035}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.21437/interspeech.2022-10465","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2022-10465","pdf_url":null,"source":{"id":"https://openalex.org/S4363604309","display_name":"Interspeech 2022","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2022","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.6700000166893005,"display_name":"Peace, Justice and strong institutions","id":"https://metadata.un.org/sdg/16"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W2358668433","https://openalex.org/W2376932109","https://openalex.org/W2382290278","https://openalex.org/W2350741829","https://openalex.org/W2130043461","https://openalex.org/W2530322880","https://openalex.org/W1596801655"],"abstract_inverted_index":{"We":[0,90],"present":[1],"a":[2,74,112,119],"streaming,":[3],"Transformer-based":[4],"end-to-end":[5,55],"automatic":[6],"speech":[7],"recognition":[8],"(ASR)":[9],"architecture":[10,21,52],"which":[11],"achieves":[12],"efficient":[13],"neural":[14],"inference":[15,28],"through":[16],"compute":[17,35,43,86,98,114],"cost":[18,115],"amortization.":[19],"Our":[20,107],"creates":[22],"sparse":[23],"computation":[24],"pathways":[25],"dynamically":[26],"at":[27,63],"time,":[29],"resulting":[30],"in":[31,42],"selective":[32],"use":[33],"of":[34,85],"resources":[36],"throughout":[37],"decoding,":[38],"enabling":[39],"significant":[40],"reductions":[41],"with":[44,56,117],"minimal":[45],"impact":[46],"on":[47,70,104],"accuracy.":[48],"The":[49],"fully":[50],"differentiable":[51],"is":[53,78],"trained":[54],"an":[57],"accompanying":[58],"lightweight":[59],"arbitrator":[60],"mechanism":[61],"operating":[62],"the":[64,82,97],"frame-level":[65],"to":[66,80],"make":[67],"dynamic":[68],"decisions":[69],"each":[71],"input":[72],"while":[73],"tunable":[75],"loss":[76],"function":[77],"used":[79],"regularize":[81],"overall":[83],"level":[84],"against":[87],"predictive":[88],"performance.":[89],"report":[91],"empirical":[92],"results":[93],"from":[94],"experiments":[95],"using":[96],"amortized":[99],"Transformer-Transducer":[100],"(T-T)":[101],"model":[102,109],"conducted":[103],"LibriSpeech":[105],"data.":[106],"best":[108],"can":[110],"achieve":[111],"60%":[113],"reduction":[116],"only":[118],"3%":[120],"relative":[121],"word":[122],"error":[123],"rate":[124],"(WER)":[125],"increase.":[126]},"counts_by_year":[{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":5}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}