{"id":"https://openalex.org/W4406858440","doi":"https://doi.org/10.1109/apsipaasc63619.2025.10848913","title":"Generalized SpecAugment: Robust Online Augmentation Technique for End-to-End Automatic Speech Recognition","display_name":"Generalized SpecAugment: Robust Online Augmentation Technique for End-to-End Automatic Speech Recognition","publication_year":2024,"publication_date":"2024-12-03","ids":{"openalex":"https://openalex.org/W4406858440","doi":"https://doi.org/10.1109/apsipaasc63619.2025.10848913"},"language":"en","primary_location":{"id":"doi:10.1109/apsipaasc63619.2025.10848913","is_oa":false,"landing_page_url":"https://doi.org/10.1109/apsipaasc63619.2025.10848913","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)","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/A5003062385","display_name":"Meet Soni","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Meet Soni","raw_affiliation_strings":["Rakuten Institute of Technology,Bangalore,India"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Rakuten Institute of Technology,Bangalore,India","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103959037","display_name":"Ashish Panda","orcid":null},"institutions":[{"id":"https://openalex.org/I55215948","display_name":"Tata Consultancy Services (India)","ror":"https://ror.org/01b9n8m42","country_code":"IN","type":"company","lineage":["https://openalex.org/I4210086519","https://openalex.org/I55215948"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Ashish Panda","raw_affiliation_strings":["TCS Research,Mumbai,India"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"TCS Research,Mumbai,India","institution_ids":["https://openalex.org/I55215948"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5047383705","display_name":"Sunil Kumar Kopparapu","orcid":"https://orcid.org/0000-0002-0502-527X"},"institutions":[{"id":"https://openalex.org/I55215948","display_name":"Tata Consultancy Services (India)","ror":"https://ror.org/01b9n8m42","country_code":"IN","type":"company","lineage":["https://openalex.org/I4210086519","https://openalex.org/I55215948"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Sunil Kumar Kopparapu","raw_affiliation_strings":["TCS Research,Mumbai,India"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"TCS Research,Mumbai,India","institution_ids":["https://openalex.org/I55215948"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.3057,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.58153585,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":95,"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/T10860","display_name":"Speech and Audio Processing","score":0.9976000189781189,"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":0.9976000189781189,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.9958000183105469,"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/T10901","display_name":"Advanced Data Compression Techniques","score":0.9527000188827515,"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/end-to-end-principle","display_name":"End-to-end principle","score":0.8677181601524353},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7181060314178467},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.597671627998352},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.4316405653953552},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3598994016647339}],"concepts":[{"id":"https://openalex.org/C74296488","wikidata":"https://www.wikidata.org/wiki/Q2527392","display_name":"End-to-end principle","level":2,"score":0.8677181601524353},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7181060314178467},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.597671627998352},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.4316405653953552},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3598994016647339},{"id":"https://openalex.org/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"score":0.0},{"id":"https://openalex.org/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"score":0.0},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/apsipaasc63619.2025.10848913","is_oa":false,"landing_page_url":"https://doi.org/10.1109/apsipaasc63619.2025.10848913","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)","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":26,"referenced_works":["https://openalex.org/W71792008","https://openalex.org/W1494198834","https://openalex.org/W2397147568","https://openalex.org/W2407080277","https://openalex.org/W2617258110","https://openalex.org/W2759921250","https://openalex.org/W2888493875","https://openalex.org/W2936774411","https://openalex.org/W2940200615","https://openalex.org/W2962780374","https://openalex.org/W2963250244","https://openalex.org/W2972389417","https://openalex.org/W2972818416","https://openalex.org/W2972963519","https://openalex.org/W2972991469","https://openalex.org/W2983434507","https://openalex.org/W3008525923","https://openalex.org/W3015726069","https://openalex.org/W3015995734","https://openalex.org/W3094833745","https://openalex.org/W3142067363","https://openalex.org/W6631362777","https://openalex.org/W6640090968","https://openalex.org/W6675409298","https://openalex.org/W6691509046","https://openalex.org/W6775557069"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2151749779","https://openalex.org/W3179968364","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W4404782863"],"abstract_inverted_index":{"Since":[0],"its":[1,32],"introduction,":[2],"SpecAugment":[3,76,122,150],"has":[4,116],"become":[5],"a":[6,72,111],"default":[7],"augmentation":[8],"technique":[9],"in":[10,48,103,127],"many":[11],"End-to-End":[12],"Automatic":[13],"Speech":[14],"Recognition":[15],"systems.":[16],"It":[17],"is":[18],"computationaly":[19],"efficient":[20],"and":[21,36,53,99,129,141,158],"provides":[22,154],"significant":[23,125],"performance":[24,46,157],"boost":[25],"without":[26],"increasing":[27],"training":[28],"time":[29,98],"due":[30],"to":[31,44],"online":[33],"nature.":[34],"Time-masking":[35],"Frequency-masking,":[37],"the":[38,42,45,51,64,93,97,104,121,162],"operations":[39],"that":[40,145],"contribute":[41],"most":[43],"gain":[47,126],"SpecAugment,":[49],"replace":[50,96],"time-stamp":[52],"certain":[54],"frequency":[55,100],"bands":[56],"with":[57,84,107,151],"either":[58],"0":[59],"or":[60],"mean":[61],"value":[62],"of":[63,92,110],"input":[65,105],"features.":[66],"In":[67,89],"this":[68],"paper,":[69],"we":[70,95],"propose":[71],"framework":[73],"called":[74],"Generalized":[75],"(Gen-SA),":[77],"where":[78],"masked":[79],"values":[80,102],"can":[81],"be":[82],"replaced":[83],"any":[85],"valid":[86],"magnitude":[87],"value.":[88],"our":[90],"implementation":[91],"Gen-SA,":[94],"mask":[101],"Mel-Spectrum":[106,109],"scaled":[108],"white":[112],"noise":[113],"signal.":[114],"Gen-SA":[115,146],"similar":[117,152],"computational":[118],"complexity":[119],"as":[120],"while":[123],"providing":[124],"robustness":[128,160],"uses":[130],"just":[131],"one":[132],"additional":[133],"signal":[134],"for":[135],"augmentation.":[136],"Experiments":[137],"on":[138],"Librispeech,":[139],"Aurora-4":[140],"TED-LIUM":[142],"datasets":[143],"show":[144],"consistently":[147],"outperforms":[148],"baseline":[149],"parameters,":[153],"better":[155],"cross-dataset":[156],"improves":[159],"against":[161],"additive":[163],"noise.":[164]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}