{"id":"https://openalex.org/W4372189025","doi":"https://doi.org/10.1109/icassp49357.2023.10094667","title":"Massively Multilingual ASR on 70 Languages: Tokenization, Architecture, and Generalization Capabilities","display_name":"Massively Multilingual ASR on 70 Languages: Tokenization, Architecture, and Generalization Capabilities","publication_year":2023,"publication_date":"2023-05-05","ids":{"openalex":"https://openalex.org/W4372189025","doi":"https://doi.org/10.1109/icassp49357.2023.10094667"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49357.2023.10094667","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49357.2023.10094667","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","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/A5038296765","display_name":"Andros Tjandra","orcid":"https://orcid.org/0000-0003-1246-5908"},"institutions":[{"id":"https://openalex.org/I4210140397","display_name":"Metrica (United States)","ror":"https://ror.org/031ffw737","country_code":"US","type":"company","lineage":["https://openalex.org/I4210140397"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Andros Tjandra","raw_affiliation_strings":["Meta,AI,USA","Meta, AI, USA"],"affiliations":[{"raw_affiliation_string":"Meta,AI,USA","institution_ids":["https://openalex.org/I4210140397"]},{"raw_affiliation_string":"Meta, AI, USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5038170826","display_name":"Nayan Singhal","orcid":"https://orcid.org/0000-0002-3189-6693"},"institutions":[{"id":"https://openalex.org/I4210140397","display_name":"Metrica (United States)","ror":"https://ror.org/031ffw737","country_code":"US","type":"company","lineage":["https://openalex.org/I4210140397"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Nayan Singhal","raw_affiliation_strings":["Meta,AI,USA","Meta, AI, USA"],"affiliations":[{"raw_affiliation_string":"Meta,AI,USA","institution_ids":["https://openalex.org/I4210140397"]},{"raw_affiliation_string":"Meta, AI, USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100325058","display_name":"David Zhang","orcid":"https://orcid.org/0000-0002-5027-5286"},"institutions":[{"id":"https://openalex.org/I4210140397","display_name":"Metrica (United States)","ror":"https://ror.org/031ffw737","country_code":"US","type":"company","lineage":["https://openalex.org/I4210140397"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"David Zhang","raw_affiliation_strings":["Meta,AI,USA","Meta, AI, USA"],"affiliations":[{"raw_affiliation_string":"Meta,AI,USA","institution_ids":["https://openalex.org/I4210140397"]},{"raw_affiliation_string":"Meta, AI, USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066166549","display_name":"Ozlem Kalinli","orcid":null},"institutions":[{"id":"https://openalex.org/I4210140397","display_name":"Metrica (United States)","ror":"https://ror.org/031ffw737","country_code":"US","type":"company","lineage":["https://openalex.org/I4210140397"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ozlem Kalinli","raw_affiliation_strings":["Meta,AI,USA","Meta, AI, USA"],"affiliations":[{"raw_affiliation_string":"Meta,AI,USA","institution_ids":["https://openalex.org/I4210140397"]},{"raw_affiliation_string":"Meta, AI, USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101170706","display_name":"Abdelrahman Mohamed","orcid":null},"institutions":[{"id":"https://openalex.org/I4210140397","display_name":"Metrica (United States)","ror":"https://ror.org/031ffw737","country_code":"US","type":"company","lineage":["https://openalex.org/I4210140397"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Abdelrahman Mohamed","raw_affiliation_strings":["Meta,AI,USA","Meta, AI, USA"],"affiliations":[{"raw_affiliation_string":"Meta,AI,USA","institution_ids":["https://openalex.org/I4210140397"]},{"raw_affiliation_string":"Meta, AI, USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5085310066","display_name":"Manh Duc Le","orcid":"https://orcid.org/0000-0003-3012-6053"},"institutions":[{"id":"https://openalex.org/I4210140397","display_name":"Metrica (United States)","ror":"https://ror.org/031ffw737","country_code":"US","type":"company","lineage":["https://openalex.org/I4210140397"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Duc Le","raw_affiliation_strings":["Meta,AI,USA","Meta, AI, USA"],"affiliations":[{"raw_affiliation_string":"Meta,AI,USA","institution_ids":["https://openalex.org/I4210140397"]},{"raw_affiliation_string":"Meta, AI, USA","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5041313589","display_name":"Michael L. Seltzer","orcid":"https://orcid.org/0000-0003-3474-2451"},"institutions":[{"id":"https://openalex.org/I4210140397","display_name":"Metrica (United States)","ror":"https://ror.org/031ffw737","country_code":"US","type":"company","lineage":["https://openalex.org/I4210140397"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Michael L. Seltzer","raw_affiliation_strings":["Meta,AI,USA","Meta, AI, USA"],"affiliations":[{"raw_affiliation_string":"Meta,AI,USA","institution_ids":["https://openalex.org/I4210140397"]},{"raw_affiliation_string":"Meta, AI, USA","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5038296765"],"corresponding_institution_ids":["https://openalex.org/I4210140397"],"apc_list":null,"apc_paid":null,"fwci":1.5459,"has_fulltext":false,"cited_by_count":9,"citation_normalized_percentile":{"value":0.85864598,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":96,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"5"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9979000091552734,"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":0.9979000091552734,"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/T10028","display_name":"Topic Modeling","score":0.9965000152587891,"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":0.9965000152587891,"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":0.8295637369155884},{"id":"https://openalex.org/keywords/embedding","display_name":"Embedding","score":0.6144235730171204},{"id":"https://openalex.org/keywords/lexical-analysis","display_name":"Lexical analysis","score":0.5884972810745239},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.46462661027908325},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4392399191856384},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.435735285282135},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.41270050406455994},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.08534923195838928}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8295637369155884},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.6144235730171204},{"id":"https://openalex.org/C176982825","wikidata":"https://www.wikidata.org/wiki/Q835922","display_name":"Lexical analysis","level":2,"score":0.5884972810745239},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.46462661027908325},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4392399191856384},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.435735285282135},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.41270050406455994},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.08534923195838928},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0},{"id":"https://openalex.org/C187736073","wikidata":"https://www.wikidata.org/wiki/Q2920921","display_name":"Management","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp49357.2023.10094667","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49357.2023.10094667","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2023 - 2023 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":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":46,"referenced_works":["https://openalex.org/W1522301498","https://openalex.org/W1828163288","https://openalex.org/W2046932483","https://openalex.org/W2064675550","https://openalex.org/W2143612262","https://openalex.org/W2327501763","https://openalex.org/W2338908902","https://openalex.org/W2407080277","https://openalex.org/W2526425061","https://openalex.org/W2763421725","https://openalex.org/W2799800213","https://openalex.org/W2899663614","https://openalex.org/W2936774411","https://openalex.org/W2962760690","https://openalex.org/W2962824709","https://openalex.org/W2962826786","https://openalex.org/W2962893195","https://openalex.org/W2963382687","https://openalex.org/W2963414781","https://openalex.org/W2963850025","https://openalex.org/W2963979492","https://openalex.org/W2971840980","https://openalex.org/W3016010032","https://openalex.org/W3095410713","https://openalex.org/W3096032230","https://openalex.org/W3096215352","https://openalex.org/W3096767609","https://openalex.org/W3162665866","https://openalex.org/W3195874849","https://openalex.org/W3213029956","https://openalex.org/W3213673948","https://openalex.org/W4210463634","https://openalex.org/W4288088457","https://openalex.org/W4372348492","https://openalex.org/W4385245566","https://openalex.org/W6631190155","https://openalex.org/W6638749077","https://openalex.org/W6703652217","https://openalex.org/W6739901393","https://openalex.org/W6745245109","https://openalex.org/W6747398299","https://openalex.org/W6755977528","https://openalex.org/W6769806307","https://openalex.org/W6780226713","https://openalex.org/W6811088048","https://openalex.org/W6843254274"],"related_works":["https://openalex.org/W4386014872","https://openalex.org/W4361193986","https://openalex.org/W1847536016","https://openalex.org/W1992967078","https://openalex.org/W2601638452","https://openalex.org/W2285263069","https://openalex.org/W4376107815","https://openalex.org/W4319309671","https://openalex.org/W4319309603","https://openalex.org/W1748623649"],"abstract_inverted_index":{"End-to-end":[0],"multilingual":[1,51,108,124],"ASR":[2,52,109,125],"has":[3],"become":[4],"more":[5],"appealing":[6],"because":[7],"of":[8,33,39,82],"several":[9],"reasons":[10],"such":[11],"as":[12],"simplifying":[13],"the":[14,31,74,80],"training":[15],"and":[16,19,37,64,66,70,99,132,136,145],"deployment":[17],"process":[18],"positive":[20],"performance":[21],"transfer":[22],"from":[23],"high-resource":[24],"to":[25,95,102,117],"low-resource":[26],"languages.":[27,56,87],"However,":[28],"scaling":[29],"up":[30],"number":[32,38],"languages,":[34],"total":[35],"hours,":[36],"unique":[40],"tokens":[41],"is":[42],"not":[43],"a":[44],"trivial":[45],"task.":[46],"This":[47],"paper":[48],"explores":[49],"large-scale":[50],"models":[53],"on":[54,128,139],"70":[55],"We":[57,120],"inspect":[58],"two":[59],"architectures:":[60],"(1)":[61],"Shared":[62],"embedding":[63,69,98],"output":[65,71,100],"(2)":[67],"Multiple":[68],"model.":[72],"In":[73],"shared":[75],"model":[76,101],"experiments,":[77],"we":[78,89],"show":[79,121],"importance":[81],"tokenization":[83,93],"strategy":[84,94],"across":[85],"different":[86],"Later,":[88],"use":[90],"our":[91,105,123],"optimal":[92],"train":[96],"multiple":[97],"further":[103],"improve":[104],"result.":[106],"Our":[107],"achieves":[110],"13.9%-15.6%":[111],"average":[112],"WER":[113,138],"relative":[114],"improvement":[115],"compared":[116],"monolingual":[118],"models.":[119],"that":[122],"generalizes":[126],"well":[127],"an":[129],"unseen":[130],"dataset":[131],"domain,":[133],"achieving":[134],"9.5%":[135],"7.5%":[137],"Multilingual":[140],"Librispeech":[141],"(MLS)":[142],"with":[143],"zero-shot":[144],"finetuning,":[146],"respectively.":[147]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":4},{"year":2024,"cited_by_count":4}],"updated_date":"2026-04-17T18:11:37.981687","created_date":"2025-10-10T00:00:00"}