{"id":"https://openalex.org/W4226120743","doi":"https://doi.org/10.21437/interspeech.2022-10937","title":"MAESTRO: Matched Speech Text Representations through Modality Matching","display_name":"MAESTRO: Matched Speech Text Representations through Modality Matching","publication_year":2022,"publication_date":"2022-09-16","ids":{"openalex":"https://openalex.org/W4226120743","doi":"https://doi.org/10.21437/interspeech.2022-10937"},"language":"en","primary_location":{"id":"doi:10.21437/interspeech.2022-10937","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2022-10937","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":"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/A5002433660","display_name":"Zhehuai Chen","orcid":"https://orcid.org/0000-0003-4400-5340"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Zhehuai Chen","raw_affiliation_strings":["Google, Inc"],"affiliations":[{"raw_affiliation_string":"Google, Inc","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100619178","display_name":"Zhang Yu","orcid":"https://orcid.org/0000-0003-2012-226X"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yu Zhang","raw_affiliation_strings":["Google, Inc"],"affiliations":[{"raw_affiliation_string":"Google, Inc","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5024913801","display_name":"Andrew E. Rosenberg","orcid":"https://orcid.org/0000-0002-4536-8824"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Andrew Rosenberg","raw_affiliation_strings":["Google, Inc"],"affiliations":[{"raw_affiliation_string":"Google, Inc","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5071715737","display_name":"Bhuvana Ramabhadran","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Bhuvana Ramabhadran","raw_affiliation_strings":["Google, Inc"],"affiliations":[{"raw_affiliation_string":"Google, Inc","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103874391","display_name":"Pedro J. Moreno","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Pedro J. Moreno","raw_affiliation_strings":["Google, Inc"],"affiliations":[{"raw_affiliation_string":"Google, Inc","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5024316712","display_name":"Ankur Bapna","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ankur Bapna","raw_affiliation_strings":["Google, Inc"],"affiliations":[{"raw_affiliation_string":"Google, Inc","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5003420204","display_name":"Heiga Zen","orcid":"https://orcid.org/0000-0002-8959-5471"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Heiga Zen","raw_affiliation_strings":["Google, Inc"],"affiliations":[{"raw_affiliation_string":"Google, Inc","institution_ids":["https://openalex.org/I1291425158"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5002433660"],"corresponding_institution_ids":["https://openalex.org/I1291425158"],"apc_list":null,"apc_paid":null,"fwci":7.068,"has_fulltext":false,"cited_by_count":68,"citation_normalized_percentile":{"value":0.97775264,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":89,"max":100},"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9986000061035156,"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":0.9986000061035156,"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":0.9955999851226807,"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":0.9883999824523926,"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":0.7536256909370422},{"id":"https://openalex.org/keywords/modality","display_name":"Modality (human\u2013computer interaction)","score":0.7091709971427917},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.6407129168510437},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5467312932014465},{"id":"https://openalex.org/keywords/matching","display_name":"Matching (statistics)","score":0.5430055260658264},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5302735567092896},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.13172924518585205},{"id":"https://openalex.org/keywords/statistics","display_name":"Statistics","score":0.06791192293167114}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7536256909370422},{"id":"https://openalex.org/C2780226545","wikidata":"https://www.wikidata.org/wiki/Q6888030","display_name":"Modality (human\u2013computer interaction)","level":2,"score":0.7091709971427917},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.6407129168510437},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5467312932014465},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.5430055260658264},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5302735567092896},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.13172924518585205},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.06791192293167114}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.21437/interspeech.2022-10937","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2022-10937","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":[{"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4","score":0.7300000190734863}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":40,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W2127836646","https://openalex.org/W2250357346","https://openalex.org/W2292087804","https://openalex.org/W2896457183","https://openalex.org/W2914120296","https://openalex.org/W2936774411","https://openalex.org/W2946200149","https://openalex.org/W2952809536","https://openalex.org/W2964012862","https://openalex.org/W2995181338","https://openalex.org/W3015280134","https://openalex.org/W3030437843","https://openalex.org/W3035390927","https://openalex.org/W3036601975","https://openalex.org/W3054645415","https://openalex.org/W3093579165","https://openalex.org/W3095410713","https://openalex.org/W3097286738","https://openalex.org/W3097777922","https://openalex.org/W3119308075","https://openalex.org/W3139878283","https://openalex.org/W3139918052","https://openalex.org/W3160799772","https://openalex.org/W3161296985","https://openalex.org/W3169064633","https://openalex.org/W3169483174","https://openalex.org/W3174724858","https://openalex.org/W3207222250","https://openalex.org/W3213029956","https://openalex.org/W4210327373","https://openalex.org/W4210690962","https://openalex.org/W4221145109","https://openalex.org/W4221155340","https://openalex.org/W4225272718","https://openalex.org/W4226033575","https://openalex.org/W4287329822","https://openalex.org/W4307023467","https://openalex.org/W4308349017","https://openalex.org/W4385245566"],"related_works":["https://openalex.org/W2385859805","https://openalex.org/W2530972254","https://openalex.org/W2374013449","https://openalex.org/W73545470","https://openalex.org/W2364381299","https://openalex.org/W2374430585","https://openalex.org/W2377397762","https://openalex.org/W2361654993","https://openalex.org/W2392886218","https://openalex.org/W3192589309"],"abstract_inverted_index":{"We":[0],"present":[1],"Maestro,":[2,102],"a":[3,49,103,154,163],"self-supervised":[4,31],"training":[5],"method":[6],"to":[7,21,36,63,106,118,180],"unify":[8],"representations":[9,41,57,109,134],"learnt":[10,58],"from":[11,17,33,42,59,87,110],"speech":[12,18,44,82],"and":[13,45,72,128,140,177],"text":[14,34,46],"modalities.Self-supervised":[15],"learning":[16,32],"signals":[19],"aims":[20],"learn":[22,107],"the":[23,28,56,67,84,90,93,144],"latent":[24,68],"structure":[25],"inherent":[26],"in":[27,66,143,167],"signal,":[29],"while":[30],"attempts":[35],"capture":[37],"lexical":[38],"information.Learning":[39],"aligned":[40,65,149],"unpaired":[43],"sequences":[47],"is":[48],"challenging":[50],"task.Previous":[51],"work":[52],"either":[53],"implicitly":[54],"enforced":[55],"these":[60,112],"two":[61,91],"modalities":[62,80,113],"be":[64],"space":[69,146],"through":[70,77,135,147],"multitasking":[71],"parameter":[73],"sharing":[74],"or":[75],"explicitly":[76],"conversion":[78],"of":[79,190],"via":[81],"synthesis.While":[83],"former":[85],"suffers":[86],"interference":[88],"between":[89],"modalities,":[92],"latter":[94],"introduces":[95],"additional":[96],"complexity.In":[97],"this":[98],"paper,":[99],"we":[100],"propose":[101],"novel":[104],"algorithm":[105],"unified":[108,133],"both":[111],"simultaneously":[114],"that":[115],"can":[116],"transfer":[117],"diverse":[119],"downstream":[120],"tasks":[121],"such":[122],"as":[123],"Automated":[124],"Speech":[125,129],"Recognition":[126],"(ASR)":[127],"Translation":[130],"(ST).Maestro":[131],"learns":[132],"sequence":[136],"alignment,":[137],"duration":[138],"prediction":[139],"matching":[141],"embeddings":[142],"learned":[145],"an":[148,188],"masked-language":[150],"model":[151],"loss.We":[152],"establish":[153],"new":[155],"state-of-the-art":[156],"(SOTA)":[157],"on":[158,184],"VoxPopuli":[159],"multilingual":[160,182],"ASR":[161,174],"with":[162,187],"8%":[164],"relative":[165],"reduction":[166],"Word":[168],"Error":[169],"Rate":[170],"(WER),":[171],"multidomain":[172],"SpeechStew":[173],"(3.7%":[175],"relative)":[176],"21":[178,195],"languages":[179],"English":[181],"ST":[183],"CoVoST":[185],"2":[186],"improvement":[189],"2.8":[191],"BLEU":[192],"averaged":[193],"over":[194],"languages.":[196]},"counts_by_year":[{"year":2025,"cited_by_count":11},{"year":2024,"cited_by_count":19},{"year":2023,"cited_by_count":37},{"year":2022,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}