{"id":"https://openalex.org/W4297841853","doi":"https://doi.org/10.21437/interspeech.2022-10961","title":"MAE-AST: Masked Autoencoding Audio Spectrogram Transformer","display_name":"MAE-AST: Masked Autoencoding Audio Spectrogram Transformer","publication_year":2022,"publication_date":"2022-09-16","ids":{"openalex":"https://openalex.org/W4297841853","doi":"https://doi.org/10.21437/interspeech.2022-10961"},"language":"en","primary_location":{"id":"doi:10.21437/interspeech.2022-10961","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2022-10961","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":"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/A5004839959","display_name":"Alan Baade","orcid":null},"institutions":[{"id":"https://openalex.org/I86519309","display_name":"The University of Texas at Austin","ror":"https://ror.org/00hj54h04","country_code":"US","type":"education","lineage":["https://openalex.org/I86519309"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Alan Baade","raw_affiliation_strings":["Department of Computer Science, The University of Texas at Austin"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Science, The University of Texas at Austin","institution_ids":["https://openalex.org/I86519309"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5075735963","display_name":"Puyuan Peng","orcid":null},"institutions":[{"id":"https://openalex.org/I86519309","display_name":"The University of Texas at Austin","ror":"https://ror.org/00hj54h04","country_code":"US","type":"education","lineage":["https://openalex.org/I86519309"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Puyuan Peng","raw_affiliation_strings":["Department of Computer Science, The University of Texas at Austin"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Science, The University of Texas at Austin","institution_ids":["https://openalex.org/I86519309"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5004717608","display_name":"David Harwath","orcid":"https://orcid.org/0000-0003-0206-0253"},"institutions":[{"id":"https://openalex.org/I86519309","display_name":"The University of Texas at Austin","ror":"https://ror.org/00hj54h04","country_code":"US","type":"education","lineage":["https://openalex.org/I86519309"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"David Harwath","raw_affiliation_strings":["Department of Computer Science, The University of Texas at Austin"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Science, The University of Texas at Austin","institution_ids":["https://openalex.org/I86519309"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I86519309"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":85,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"2438","last_page":"2442"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10860","display_name":"Speech and Audio Processing","score":0.887499988079071,"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"}},"topics":[{"id":"https://openalex.org/T10860","display_name":"Speech and Audio Processing","score":0.887499988079071,"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/T10688","display_name":"Image and Signal Denoising Methods","score":0.8349000215530396,"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/T10901","display_name":"Advanced Data Compression Techniques","score":0.7422000169754028,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/spectrogram","display_name":"Spectrogram","score":0.9339027404785156},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6488469243049622},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.6054911017417908},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5294432640075684},{"id":"https://openalex.org/keywords/electrical-engineering","display_name":"Electrical engineering","score":0.13126343488693237},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.10961750149726868},{"id":"https://openalex.org/keywords/voltage","display_name":"Voltage","score":0.07214784622192383}],"concepts":[{"id":"https://openalex.org/C45273575","wikidata":"https://www.wikidata.org/wiki/Q578970","display_name":"Spectrogram","level":2,"score":0.9339027404785156},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6488469243049622},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.6054911017417908},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5294432640075684},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.13126343488693237},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.10961750149726868},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.07214784622192383}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.21437/interspeech.2022-10961","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2022-10961","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":"Affordable and clean energy","score":0.7900000214576721,"id":"https://metadata.un.org/sdg/7"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":21,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W1522301498","https://openalex.org/W2146334809","https://openalex.org/W2593116425","https://openalex.org/W2797583228","https://openalex.org/W2896457183","https://openalex.org/W2933138175","https://openalex.org/W2981087920","https://openalex.org/W3036601975","https://openalex.org/W3094502228","https://openalex.org/W3138516171","https://openalex.org/W3160799772","https://openalex.org/W3196974791","https://openalex.org/W3197580070","https://openalex.org/W3206996142","https://openalex.org/W4287667694","https://openalex.org/W4292779060","https://openalex.org/W4297808394","https://openalex.org/W4313156423","https://openalex.org/W4385245566","https://openalex.org/W4398958419"],"related_works":["https://openalex.org/W2973062255","https://openalex.org/W2897924318","https://openalex.org/W2138997758","https://openalex.org/W2766960583","https://openalex.org/W2065296656","https://openalex.org/W3195104037","https://openalex.org/W3133517635","https://openalex.org/W2395582662","https://openalex.org/W2900550435","https://openalex.org/W4310471687"],"abstract_inverted_index":{"In":[0],"this":[1,54],"paper,":[2],"we":[3,24,129],"propose":[4],"a":[5,32,72,81,97,138],"simple":[6],"yet":[7],"powerful":[8],"improvement":[9],"over":[10,105],"the":[11,26,29,42,57,69,106,127,135,159],"recent":[12],"Self-Supervised":[13],"Audio":[14],"Spectrogram":[15],"Transformer":[16],"(SSAST)":[17],"model":[18,116],"for":[19],"speech":[20],"and":[21,80,88,100,117,152,161],"audio":[22,111,162],"classification.Specifically,":[23],"leverage":[25],"insight":[27],"that":[28,41,92,131],"SSAST":[30,108,136],"uses":[31,126],"very":[33],"high":[34],"masking":[35],"ratio":[36],"(75%)":[37],"during":[38],"pretraining,":[39],"meaning":[40],"vast":[43],"majority":[44],"of":[45,140,150],"self-attention":[46],"compute":[47],"is":[48],"performed":[49],"on":[50,76,85,121,137],"mask":[51,89],"tokens.We":[52,90],"address":[53],"by":[55],"integrating":[56],"encoder-decoder":[58],"architecture":[59],"from":[60],"Masked":[61],"Autoencoders":[62],"are":[63],"Scalable":[64],"Vision":[65],"Learners":[66],"(MAE)":[67],"into":[68,147],"SSAST,":[70],"where":[71],"deep":[73],"encoder":[74,86],"operates":[75,84],"only":[77,125],"unmasked":[78],"input,":[79],"shallow":[82],"decoder":[83],"outputs":[87],"find":[91,130],"MAE-like":[93],"pretraining":[94,112,151,157],"can":[95],"provide":[96],"3\u00d7":[98],"speedup":[99],"2\u00d7":[101],"memory":[102],"usage":[103],"reduction":[104],"vanilla":[107],"using":[109],"current":[110],"strategies":[113,149],"with":[114],"ordinary":[115],"input":[118],"sizes.When":[119],"finetuning":[120],"downstream":[122,141],"tasks,":[123],"which":[124],"encoder,":[128],"our":[132],"approach":[133],"outperforms":[134],"variety":[139],"tasks.We":[142],"further":[143],"conduct":[144],"comprehensive":[145],"evaluations":[146],"different":[148],"explore":[153],"differences":[154],"in":[155],"MAE-style":[156],"between":[158],"visual":[160],"domains.":[163]},"counts_by_year":[{"year":2026,"cited_by_count":7},{"year":2025,"cited_by_count":25},{"year":2024,"cited_by_count":32},{"year":2023,"cited_by_count":19},{"year":2022,"cited_by_count":2}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}