{"id":"https://openalex.org/W4319862449","doi":"https://doi.org/10.1109/slt54892.2023.10022592","title":"Accelerator-Aware Training for Transducer-Based Speech Recognition","display_name":"Accelerator-Aware Training for Transducer-Based Speech Recognition","publication_year":2023,"publication_date":"2023-01-09","ids":{"openalex":"https://openalex.org/W4319862449","doi":"https://doi.org/10.1109/slt54892.2023.10022592"},"language":"en","primary_location":{"id":"doi:10.1109/slt54892.2023.10022592","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/slt54892.2023.10022592","pdf_url":null,"source":{"id":"https://openalex.org/S4363605953","display_name":"2022 IEEE Spoken Language Technology Workshop (SLT)","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":"2022 IEEE Spoken Language Technology Workshop (SLT)","raw_type":"proceedings-article"},"type":"article","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2305.07778","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5062889462","display_name":"Suhaila Shakiah","orcid":"https://orcid.org/0000-0002-1891-7058"},"institutions":[{"id":"https://openalex.org/I1311688040","display_name":"Amazon (United States)","ror":"https://ror.org/04mv4n011","country_code":"US","type":"company","lineage":["https://openalex.org/I1311688040"]},{"id":"https://openalex.org/I49313256","display_name":"Alexion Pharma (Switzerland)","ror":"https://ror.org/05hkjyh45","country_code":"CH","type":"company","lineage":["https://openalex.org/I49313256"]}],"countries":["CH","US"],"is_corresponding":false,"raw_author_name":"Suhaila M. Shakiah","raw_affiliation_strings":["Alexa Machine Learning, Amazon.com,USA","Alexa Machine Learning, Amazon.com, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alexa Machine Learning, Amazon.com,USA","institution_ids":["https://openalex.org/I1311688040","https://openalex.org/I49313256"]},{"raw_affiliation_string":"Alexa Machine Learning, Amazon.com, USA","institution_ids":["https://openalex.org/I1311688040"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5076547892","display_name":"Rupak Vignesh Swaminathan","orcid":null},"institutions":[{"id":"https://openalex.org/I1311688040","display_name":"Amazon (United States)","ror":"https://ror.org/04mv4n011","country_code":"US","type":"company","lineage":["https://openalex.org/I1311688040"]},{"id":"https://openalex.org/I49313256","display_name":"Alexion Pharma (Switzerland)","ror":"https://ror.org/05hkjyh45","country_code":"CH","type":"company","lineage":["https://openalex.org/I49313256"]}],"countries":["CH","US"],"is_corresponding":false,"raw_author_name":"Rupak Vignesh Swaminathan","raw_affiliation_strings":["Alexa Machine Learning, Amazon.com,USA","Alexa Machine Learning, Amazon.com, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alexa Machine Learning, Amazon.com,USA","institution_ids":["https://openalex.org/I1311688040","https://openalex.org/I49313256"]},{"raw_affiliation_string":"Alexa Machine Learning, Amazon.com, USA","institution_ids":["https://openalex.org/I1311688040"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101663835","display_name":"Hieu Duy Nguyen","orcid":"https://orcid.org/0000-0002-8774-5122"},"institutions":[{"id":"https://openalex.org/I1311688040","display_name":"Amazon (United States)","ror":"https://ror.org/04mv4n011","country_code":"US","type":"company","lineage":["https://openalex.org/I1311688040"]},{"id":"https://openalex.org/I49313256","display_name":"Alexion Pharma (Switzerland)","ror":"https://ror.org/05hkjyh45","country_code":"CH","type":"company","lineage":["https://openalex.org/I49313256"]}],"countries":["CH","US"],"is_corresponding":false,"raw_author_name":"Hieu Duy Nguyen","raw_affiliation_strings":["Alexa Machine Learning, Amazon.com,USA","Alexa Machine Learning, Amazon.com, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alexa Machine Learning, Amazon.com,USA","institution_ids":["https://openalex.org/I1311688040","https://openalex.org/I49313256"]},{"raw_affiliation_string":"Alexa Machine Learning, Amazon.com, USA","institution_ids":["https://openalex.org/I1311688040"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066216178","display_name":"Raviteja Chinta","orcid":null},"institutions":[{"id":"https://openalex.org/I1311688040","display_name":"Amazon (United States)","ror":"https://ror.org/04mv4n011","country_code":"US","type":"company","lineage":["https://openalex.org/I1311688040"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Raviteja Chinta","raw_affiliation_strings":["Amazon.com,Hardware Compute Group,USA","Hardware Compute Group, Amazon.com, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Amazon.com,Hardware Compute Group,USA","institution_ids":["https://openalex.org/I1311688040"]},{"raw_affiliation_string":"Hardware Compute Group, Amazon.com, USA","institution_ids":["https://openalex.org/I1311688040"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5076634486","display_name":"Tariq Afzal","orcid":"https://orcid.org/0000-0002-9922-1385"},"institutions":[{"id":"https://openalex.org/I1311688040","display_name":"Amazon (United States)","ror":"https://ror.org/04mv4n011","country_code":"US","type":"company","lineage":["https://openalex.org/I1311688040"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Tariq Afzal","raw_affiliation_strings":["Amazon.com,Hardware Compute Group,USA","Alexa Machine Learning, Amazon.com, USA","Hardware Compute Group, Amazon.com, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Amazon.com,Hardware Compute Group,USA","institution_ids":["https://openalex.org/I1311688040"]},{"raw_affiliation_string":"Alexa Machine Learning, Amazon.com, USA","institution_ids":["https://openalex.org/I1311688040"]},{"raw_affiliation_string":"Hardware Compute Group, Amazon.com, USA","institution_ids":["https://openalex.org/I1311688040"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5062762746","display_name":"Nathan Susanj","orcid":null},"institutions":[{"id":"https://openalex.org/I1311688040","display_name":"Amazon (United States)","ror":"https://ror.org/04mv4n011","country_code":"US","type":"company","lineage":["https://openalex.org/I1311688040"]},{"id":"https://openalex.org/I49313256","display_name":"Alexion Pharma (Switzerland)","ror":"https://ror.org/05hkjyh45","country_code":"CH","type":"company","lineage":["https://openalex.org/I49313256"]}],"countries":["CH","US"],"is_corresponding":false,"raw_author_name":"Nathan Susanj","raw_affiliation_strings":["Alexa Machine Learning, Amazon.com,USA","Alexa Machine Learning, Amazon.com, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alexa Machine Learning, Amazon.com,USA","institution_ids":["https://openalex.org/I1311688040","https://openalex.org/I49313256"]},{"raw_affiliation_string":"Alexa Machine Learning, Amazon.com, USA","institution_ids":["https://openalex.org/I1311688040"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5029573067","display_name":"Athanasios Mouchtaris","orcid":"https://orcid.org/0000-0001-7583-0189"},"institutions":[{"id":"https://openalex.org/I1311688040","display_name":"Amazon (United States)","ror":"https://ror.org/04mv4n011","country_code":"US","type":"company","lineage":["https://openalex.org/I1311688040"]},{"id":"https://openalex.org/I49313256","display_name":"Alexion Pharma (Switzerland)","ror":"https://ror.org/05hkjyh45","country_code":"CH","type":"company","lineage":["https://openalex.org/I49313256"]}],"countries":["CH","US"],"is_corresponding":false,"raw_author_name":"Athanasios Mouchtaris","raw_affiliation_strings":["Alexa Machine Learning, Amazon.com,USA","Alexa Machine Learning, Amazon.com, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alexa Machine Learning, Amazon.com,USA","institution_ids":["https://openalex.org/I1311688040","https://openalex.org/I49313256"]},{"raw_affiliation_string":"Alexa Machine Learning, Amazon.com, USA","institution_ids":["https://openalex.org/I1311688040"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5023708721","display_name":"Grant P. Strimel","orcid":null},"institutions":[{"id":"https://openalex.org/I1311688040","display_name":"Amazon (United States)","ror":"https://ror.org/04mv4n011","country_code":"US","type":"company","lineage":["https://openalex.org/I1311688040"]},{"id":"https://openalex.org/I49313256","display_name":"Alexion Pharma (Switzerland)","ror":"https://ror.org/05hkjyh45","country_code":"CH","type":"company","lineage":["https://openalex.org/I49313256"]}],"countries":["CH","US"],"is_corresponding":false,"raw_author_name":"Grant P. Strimel","raw_affiliation_strings":["Alexa Machine Learning, Amazon.com,USA","Alexa Machine Learning, Amazon.com, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alexa Machine Learning, Amazon.com,USA","institution_ids":["https://openalex.org/I1311688040","https://openalex.org/I49313256"]},{"raw_affiliation_string":"Alexa Machine Learning, Amazon.com, USA","institution_ids":["https://openalex.org/I1311688040"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5110230355","display_name":"Ariya Rastrow","orcid":null},"institutions":[{"id":"https://openalex.org/I1311688040","display_name":"Amazon (United States)","ror":"https://ror.org/04mv4n011","country_code":"US","type":"company","lineage":["https://openalex.org/I1311688040"]},{"id":"https://openalex.org/I49313256","display_name":"Alexion Pharma (Switzerland)","ror":"https://ror.org/05hkjyh45","country_code":"CH","type":"company","lineage":["https://openalex.org/I49313256"]}],"countries":["CH","US"],"is_corresponding":false,"raw_author_name":"Ariya Rastrow","raw_affiliation_strings":["Alexa Machine Learning, Amazon.com,USA","Alexa Machine Learning, Amazon.com, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alexa Machine Learning, Amazon.com,USA","institution_ids":["https://openalex.org/I1311688040","https://openalex.org/I49313256"]},{"raw_affiliation_string":"Alexa Machine Learning, Amazon.com, USA","institution_ids":["https://openalex.org/I1311688040"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.1031,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":{"value":0.19773429,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":94},"biblio":{"volume":"30","issue":null,"first_page":"100","last_page":"107"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9980000257492065,"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":0.9980000257492065,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.9979000091552734,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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":0.9973000288009644,"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":0.8562135100364685},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5461682677268982},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5347563028335571},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.5091637969017029},{"id":"https://openalex.org/keywords/leverage","display_name":"Leverage (statistics)","score":0.49685075879096985},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.49498289823532104},{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.4732692241668701},{"id":"https://openalex.org/keywords/recurrent-neural-network","display_name":"Recurrent neural network","score":0.46300721168518066},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3590978980064392},{"id":"https://openalex.org/keywords/computer-vision","display_name":"Computer vision","score":0.12381798028945923}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8562135100364685},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5461682677268982},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5347563028335571},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.5091637969017029},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.49685075879096985},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.49498289823532104},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.4732692241668701},{"id":"https://openalex.org/C147168706","wikidata":"https://www.wikidata.org/wiki/Q1457734","display_name":"Recurrent neural network","level":3,"score":0.46300721168518066},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3590978980064392},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.12381798028945923},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/slt54892.2023.10022592","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/slt54892.2023.10022592","pdf_url":null,"source":{"id":"https://openalex.org/S4363605953","display_name":"2022 IEEE Spoken Language Technology Workshop (SLT)","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":"2022 IEEE Spoken Language Technology Workshop (SLT)","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2305.07778","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2305.07778","pdf_url":"https://arxiv.org/pdf/2305.07778","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"public-domain","license_id":"https://openalex.org/licenses/public-domain","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2305.07778","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2305.07778","pdf_url":"https://arxiv.org/pdf/2305.07778","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"public-domain","license_id":"https://openalex.org/licenses/public-domain","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":false},"content_urls":{"pdf":"https://content.openalex.org/works/W4319862449.pdf"},"referenced_works_count":36,"referenced_works":["https://openalex.org/W1689711448","https://openalex.org/W1828163288","https://openalex.org/W2064675550","https://openalex.org/W2119144962","https://openalex.org/W2142343760","https://openalex.org/W2150380720","https://openalex.org/W2167425979","https://openalex.org/W2242818861","https://openalex.org/W2497335382","https://openalex.org/W2764043458","https://openalex.org/W2798275680","https://openalex.org/W2892009249","https://openalex.org/W2963122961","https://openalex.org/W2963174729","https://openalex.org/W2975044525","https://openalex.org/W2981857663","https://openalex.org/W3097522836","https://openalex.org/W3097973766","https://openalex.org/W3118708382","https://openalex.org/W3151287998","https://openalex.org/W3162649911","https://openalex.org/W3163062345","https://openalex.org/W3163907627","https://openalex.org/W3170968215","https://openalex.org/W3196783077","https://openalex.org/W3197201410","https://openalex.org/W3198715852","https://openalex.org/W4297841536","https://openalex.org/W6638749077","https://openalex.org/W6682044802","https://openalex.org/W6745148473","https://openalex.org/W6750647084","https://openalex.org/W6754450847","https://openalex.org/W6768205276","https://openalex.org/W6784856619","https://openalex.org/W6788325135"],"related_works":["https://openalex.org/W4298287631","https://openalex.org/W2953061907","https://openalex.org/W3032952384","https://openalex.org/W3034302643","https://openalex.org/W1847088711","https://openalex.org/W3008625068","https://openalex.org/W4379251913","https://openalex.org/W3128807919","https://openalex.org/W3176411177","https://openalex.org/W3035501883"],"abstract_inverted_index":{"Machine":[0],"learning":[1],"model":[2],"weights":[3],"and":[4,37,113,122],"activations":[5],"are":[6],"represented":[7],"in":[8,17,62,127,137],"full-precision":[9],"during":[10,47],"training.":[11],"This":[12],"leads":[13],"to":[14,33,56,75,80,97,133],"performance":[15],"degradation":[16,54,136],"runtime":[18,35],"when":[19],"deployed":[20],"on":[21,59,116],"neural":[22],"network":[23],"accelerator":[24],"(NNA)":[25],"chips,":[26],"which":[27],"leverage":[28],"highly":[29],"parallelized":[30],"fixed-point":[31],"arithmetic":[32],"improve":[34],"memory":[36],"latency.":[38],"In":[39],"this":[40],"work,":[41],"we":[42],"replicate":[43],"the":[44,48,53,60,73,81,88],"NNA":[45,61,69],"operators":[46],"training":[49],"phase,":[50],"accounting":[51],"for":[52,105],"due":[55],"low-precision":[57],"inference":[58],"back-propagation.":[63],"Our":[64],"proposed":[65],"method":[66],"efficiently":[67],"emulates":[68],"operations,":[70],"thus":[71],"foregoing":[72],"need":[74],"transfer":[76],"quantization":[77],"error-prone":[78],"data":[79,121],"Central":[82],"Processing":[83],"Unit":[84],"(CPU),":[85],"ultimately":[86],"reducing":[87],"user":[89],"perceived":[90],"latency":[91,129],"(UPL).":[92],"We":[93,111],"apply":[94],"our":[95],"approach":[96],"Recurrent":[98],"Neural":[99],"Network-Transducer":[100],"(RNN-T),":[101],"an":[102],"attractive":[103],"architecture":[104],"on-device":[106],"streaming":[107],"speech":[108],"recognition":[109],"tasks.":[110],"train":[112],"evaluate":[114],"models":[115],"270K":[117],"hours":[118],"of":[119],"English":[120],"show":[123],"a":[124],"5-7%":[125],"improvement":[126],"engine":[128],"while":[130],"saving":[131],"up":[132],"10%":[134],"relative":[135],"WER.":[138]},"counts_by_year":[{"year":2024,"cited_by_count":1}],"updated_date":"2026-07-02T09:51:11.867554","created_date":"2025-10-10T00:00:00"}
