{"id":"https://openalex.org/W3151287998","doi":"https://doi.org/10.1109/icassp39728.2021.9414652","title":"Bifocal Neural ASR: Exploiting Keyword Spotting for Inference Optimization","display_name":"Bifocal Neural ASR: Exploiting Keyword Spotting for Inference Optimization","publication_year":2021,"publication_date":"2021-05-13","ids":{"openalex":"https://openalex.org/W3151287998","doi":"https://doi.org/10.1109/icassp39728.2021.9414652","mag":"3151287998"},"language":"en","primary_location":{"id":"doi:10.1109/icassp39728.2021.9414652","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp39728.2021.9414652","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2021 - 2021 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/A5056064437","display_name":"Jon Macoskey","orcid":null},"institutions":[{"id":"https://openalex.org/I4210089985","display_name":"Amazon (Germany)","ror":"https://ror.org/00b9ktm87","country_code":"DE","type":"company","lineage":["https://openalex.org/I1311688040","https://openalex.org/I4210089985"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Jon Macoskey","raw_affiliation_strings":["Amazon.com"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Amazon.com","institution_ids":["https://openalex.org/I4210089985"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5023708721","display_name":"Grant P. Strimel","orcid":null},"institutions":[{"id":"https://openalex.org/I4210089985","display_name":"Amazon (Germany)","ror":"https://ror.org/00b9ktm87","country_code":"DE","type":"company","lineage":["https://openalex.org/I1311688040","https://openalex.org/I4210089985"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Grant P. Strimel","raw_affiliation_strings":["Amazon.com"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Amazon.com","institution_ids":["https://openalex.org/I4210089985"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5110230355","display_name":"Ariya Rastrow","orcid":null},"institutions":[{"id":"https://openalex.org/I4210089985","display_name":"Amazon (Germany)","ror":"https://ror.org/00b9ktm87","country_code":"DE","type":"company","lineage":["https://openalex.org/I1311688040","https://openalex.org/I4210089985"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Ariya Rastrow","raw_affiliation_strings":["Amazon.com"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Amazon.com","institution_ids":["https://openalex.org/I4210089985"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I4210089985"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":14,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"5999","last_page":"6003"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9998999834060669,"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":0.9998999834060669,"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/T11309","display_name":"Music and Audio Processing","score":0.9987999796867371,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9968000054359436,"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":0.8656425476074219},{"id":"https://openalex.org/keywords/recurrent-neural-network","display_name":"Recurrent neural network","score":0.7946390509605408},{"id":"https://openalex.org/keywords/keyword-spotting","display_name":"Keyword spotting","score":0.7371466755867004},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.6647921800613403},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6645709276199341},{"id":"https://openalex.org/keywords/leverage","display_name":"Leverage (statistics)","score":0.5517162084579468},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.519950270652771},{"id":"https://openalex.org/keywords/spotting","display_name":"Spotting","score":0.49828457832336426},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4593508541584015},{"id":"https://openalex.org/keywords/architecture","display_name":"Architecture","score":0.4275456368923187},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.4212404787540436},{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.41569918394088745},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.3832632005214691},{"id":"https://openalex.org/keywords/computer-vision","display_name":"Computer vision","score":0.17962467670440674}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8656425476074219},{"id":"https://openalex.org/C147168706","wikidata":"https://www.wikidata.org/wiki/Q1457734","display_name":"Recurrent neural network","level":3,"score":0.7946390509605408},{"id":"https://openalex.org/C2781213101","wikidata":"https://www.wikidata.org/wiki/Q6398558","display_name":"Keyword spotting","level":2,"score":0.7371466755867004},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.6647921800613403},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6645709276199341},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.5517162084579468},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.519950270652771},{"id":"https://openalex.org/C2779506182","wikidata":"https://www.wikidata.org/wiki/Q7580141","display_name":"Spotting","level":2,"score":0.49828457832336426},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4593508541584015},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.4275456368923187},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.4212404787540436},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.41569918394088745},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.3832632005214691},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.17962467670440674},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"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/C153349607","wikidata":"https://www.wikidata.org/wiki/Q36649","display_name":"Visual arts","level":1,"score":0.0},{"id":"https://openalex.org/C142362112","wikidata":"https://www.wikidata.org/wiki/Q735","display_name":"Art","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp39728.2021.9414652","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp39728.2021.9414652","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4","score":0.4699999988079071}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":34,"referenced_works":["https://openalex.org/W1689711448","https://openalex.org/W1828163288","https://openalex.org/W1855892484","https://openalex.org/W2064675550","https://openalex.org/W2102113734","https://openalex.org/W2127141656","https://openalex.org/W2143612262","https://openalex.org/W2293829681","https://openalex.org/W2325237720","https://openalex.org/W2497335382","https://openalex.org/W2545177271","https://openalex.org/W2556046966","https://openalex.org/W2602634800","https://openalex.org/W2748659049","https://openalex.org/W2764043458","https://openalex.org/W2803570545","https://openalex.org/W2889129739","https://openalex.org/W2953219395","https://openalex.org/W2962760690","https://openalex.org/W2963174729","https://openalex.org/W2963250244","https://openalex.org/W2975044525","https://openalex.org/W3007227084","https://openalex.org/W3015194534","https://openalex.org/W3097522836","https://openalex.org/W6638749077","https://openalex.org/W6639156005","https://openalex.org/W6675365184","https://openalex.org/W6701498492","https://openalex.org/W6730040086","https://openalex.org/W6743881753","https://openalex.org/W6745148473","https://openalex.org/W6751832336","https://openalex.org/W6768205276"],"related_works":["https://openalex.org/W2918559346","https://openalex.org/W3119978414","https://openalex.org/W2114097550","https://openalex.org/W2516975559","https://openalex.org/W2545741539","https://openalex.org/W3206647229","https://openalex.org/W4286904253","https://openalex.org/W2000885660","https://openalex.org/W1969408022","https://openalex.org/W2117995638"],"abstract_inverted_index":{"We":[0,102],"present":[1,103],"Bifocal":[2,67,121],"RNN-T,":[3],"a":[4,28,52,61,136],"new":[5],"variant":[6],"of":[7,39,46],"the":[8,47,66,74,104],"Recurrent":[9],"Neural":[10],"Network":[11],"Transducer":[12],"(RNN-T)":[13],"architecture":[14,26,77,105],"designed":[15],"for":[16,31,51,96],"improved":[17],"inference":[18,125],"time":[19],"latency":[20],"on":[21,111],"speech":[22,99,113],"recognition":[23,100,114],"tasks.":[24,115],"The":[25,76],"enables":[27],"dynamic":[29],"pivot":[30],"its":[32],"runtime":[33],"compute":[34],"pathway,":[35],"namely":[36],"taking":[37],"advantage":[38],"keyword":[40],"spotting":[41],"to":[42,49],"select":[43],"which":[44,70],"component":[45],"network":[48],"execute":[50],"given":[53],"audio":[54],"frame.":[55],"To":[56],"accomplish":[57],"this,":[58],"we":[59,64,71,117],"leverage":[60],"recurrent":[62],"cell":[63],"call":[65],"LSTM":[68],"(BF-LSTM),":[69],"detail":[72],"in":[73,139],"paper.":[75],"is":[78],"compatible":[79],"with":[80,129],"other":[81],"optimization":[82],"strategies":[83],"such":[84],"as":[85],"quantization,":[86],"sparsification,":[87],"and":[88,106,134],"applying":[89],"time-reduction":[90],"layers,":[91],"making":[92],"it":[93],"especially":[94],"applicable":[95],"deployed,":[97],"real-time":[98],"settings.":[101],"report":[107],"comparative":[108],"experimental":[109],"results":[110],"voice-assistant":[112],"Specifically,":[116],"show":[118],"our":[119],"proposed":[120],"RNN-T":[122],"can":[123],"improve":[124],"cost":[126],"by":[127],"29.1%":[128],"matching":[130],"word":[131],"error":[132],"rates":[133],"only":[135],"minor":[137],"increase":[138],"memory":[140],"size.":[141]},"counts_by_year":[{"year":2023,"cited_by_count":6},{"year":2022,"cited_by_count":2},{"year":2021,"cited_by_count":6}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}