{"id":"https://openalex.org/W4319862667","doi":"https://doi.org/10.1109/slt54892.2023.10022937","title":"Exploration of Language-Specific Self-Attention Parameters for Multilingual End-to-End Speech Recognition","display_name":"Exploration of Language-Specific Self-Attention Parameters for Multilingual End-to-End Speech Recognition","publication_year":2023,"publication_date":"2023-01-09","ids":{"openalex":"https://openalex.org/W4319862667","doi":"https://doi.org/10.1109/slt54892.2023.10022937"},"language":"en","primary_location":{"id":"doi:10.1109/slt54892.2023.10022937","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/slt54892.2023.10022937","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":["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/A5019745645","display_name":"B. Houston","orcid":"https://orcid.org/0000-0001-5666-3629"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Brady Houston","raw_affiliation_strings":["AWS AI Labs"],"affiliations":[{"raw_affiliation_string":"AWS AI Labs","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5050325468","display_name":"Katrin Kirchhoff","orcid":"https://orcid.org/0000-0002-6645-6030"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Katrin Kirchhoff","raw_affiliation_strings":["AWS AI Labs"],"affiliations":[{"raw_affiliation_string":"AWS AI Labs","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5019745645"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.3391,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.45393024,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":96},"biblio":{"volume":"34","issue":null,"first_page":"755","last_page":"762"},"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/T12031","display_name":"Speech and dialogue systems","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/T10181","display_name":"Natural Language Processing Techniques","score":0.9977999925613403,"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":0.8554393649101257},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.7180787920951843},{"id":"https://openalex.org/keywords/pooling","display_name":"Pooling","score":0.6208236813545227},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5506526827812195},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.5434226989746094},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5343600511550903},{"id":"https://openalex.org/keywords/phone","display_name":"Phone","score":0.5273262858390808},{"id":"https://openalex.org/keywords/end-to-end-principle","display_name":"End-to-end principle","score":0.5105594396591187},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.450288325548172},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.40722382068634033},{"id":"https://openalex.org/keywords/linguistics","display_name":"Linguistics","score":0.16553115844726562}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8554393649101257},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.7180787920951843},{"id":"https://openalex.org/C70437156","wikidata":"https://www.wikidata.org/wiki/Q7228652","display_name":"Pooling","level":2,"score":0.6208236813545227},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5506526827812195},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5434226989746094},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5343600511550903},{"id":"https://openalex.org/C2778707766","wikidata":"https://www.wikidata.org/wiki/Q202064","display_name":"Phone","level":2,"score":0.5273262858390808},{"id":"https://openalex.org/C74296488","wikidata":"https://www.wikidata.org/wiki/Q2527392","display_name":"End-to-end principle","level":2,"score":0.5105594396591187},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.450288325548172},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.40722382068634033},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.16553115844726562},{"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/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/slt54892.2023.10022937","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/slt54892.2023.10022937","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"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Quality Education","score":0.7699999809265137,"id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":24,"referenced_works":["https://openalex.org/W2025198378","https://openalex.org/W2526425061","https://openalex.org/W2786835190","https://openalex.org/W2807535859","https://openalex.org/W2964309797","https://openalex.org/W2971840980","https://openalex.org/W2972619908","https://openalex.org/W3015877095","https://openalex.org/W3094667432","https://openalex.org/W3095311338","https://openalex.org/W3096032230","https://openalex.org/W3096215352","https://openalex.org/W3097338456","https://openalex.org/W3097777922","https://openalex.org/W3156323585","https://openalex.org/W3175948724","https://openalex.org/W3178647810","https://openalex.org/W3196757209","https://openalex.org/W3197932033","https://openalex.org/W3202167223","https://openalex.org/W4210463634","https://openalex.org/W4297841598","https://openalex.org/W6752241964","https://openalex.org/W6797623589"],"related_works":["https://openalex.org/W3105216938","https://openalex.org/W3209371554","https://openalex.org/W4300537895","https://openalex.org/W3197304116","https://openalex.org/W2937649809","https://openalex.org/W3184187848","https://openalex.org/W2782005958","https://openalex.org/W4285757700","https://openalex.org/W4319431239","https://openalex.org/W4287816637"],"abstract_inverted_index":{"In":[0],"the":[1,12,66,71,82,96],"last":[2],"several":[3,49],"years,":[4],"end-to-end":[5],"(E2E)":[6],"ASR":[7,16],"models":[8],"have":[9],"mostly":[10],"surpassed":[11],"performance":[13,101],"of":[14,85],"hybrid":[15],"models.":[17,88],"E2E":[18,40],"is":[19],"particularly":[20],"well":[21,61],"suited":[22],"to":[23,48,70,74,107,110,116,119],"multilingual":[24,39,112],"approaches":[25],"because":[26],"it":[27],"doesn't":[28],"require":[29],"language-specific":[30,55,79,93],"phone":[31],"alignments":[32],"for":[33],"training.":[34],"Recent":[35],"work":[36,76],"has":[37],"improved":[38],"modeling":[41],"over":[42],"naive":[43],"data":[44],"pooling":[45],"on":[46],"up":[47,106,115],"dozen":[50],"languages":[51,104],"by":[52,105],"using":[53,92],"both":[54],"and":[56,114],"language-universal":[57],"model":[58],"parameters,":[59],"as":[60,62],"providing":[63],"information":[64],"about":[65],"language":[67],"being":[68],"presented":[69],"network.":[72],"Complementary":[73],"previous":[75],"we":[77],"analyze":[78],"parameters":[80,94,126],"in":[81,95],"attention":[83,97],"mechanism":[84,98],"Conformer-based":[86],"encoder":[87],"We":[89],"show":[90],"that":[91],"can":[99],"improve":[100],"across":[102],"six":[103],"12%":[108],"compared":[109,118],"standard":[111],"baselines":[113],"36%":[117],"monolingual":[120,128],"baselines,":[121],"without":[122],"requiring":[123],"any":[124],"additional":[125],"during":[127],"inference":[129],"nor":[130],"fine-tuning.":[131]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2023,"cited_by_count":2}],"updated_date":"2025-12-21T23:12:01.093139","created_date":"2025-10-10T00:00:00"}