{"id":"https://openalex.org/W3163221127","doi":"https://doi.org/10.1109/icassp39728.2021.9414648","title":"Focus on the Present: A Regularization Method for the ASR Source-Target Attention Layer","display_name":"Focus on the Present: A Regularization Method for the ASR Source-Target Attention Layer","publication_year":2021,"publication_date":"2021-05-13","ids":{"openalex":"https://openalex.org/W3163221127","doi":"https://doi.org/10.1109/icassp39728.2021.9414648","mag":"3163221127"},"language":"en","primary_location":{"id":"doi:10.1109/icassp39728.2021.9414648","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp39728.2021.9414648","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2021 - 2021 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/A5071702566","display_name":"Nanxin Chen","orcid":"https://orcid.org/0000-0001-6698-1604"},"institutions":[{"id":"https://openalex.org/I145311948","display_name":"Johns Hopkins University","ror":"https://ror.org/00za53h95","country_code":"US","type":"education","lineage":["https://openalex.org/I145311948"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Nanxin Chen","raw_affiliation_strings":["Center for Language and Speech Processing, Johns Hopkins University, Baltimore, MD"],"affiliations":[{"raw_affiliation_string":"Center for Language and Speech Processing, Johns Hopkins University, Baltimore, MD","institution_ids":["https://openalex.org/I145311948"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5027217976","display_name":"Piotr \u017belasko","orcid":"https://orcid.org/0000-0002-8245-0413"},"institutions":[{"id":"https://openalex.org/I145311948","display_name":"Johns Hopkins University","ror":"https://ror.org/00za53h95","country_code":"US","type":"education","lineage":["https://openalex.org/I145311948"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Piotr Zelasko","raw_affiliation_strings":["Center for Language and Speech Processing, Johns Hopkins University, Baltimore, MD","Human Language Technology Center of Excellence, Johns Hopkins University, Baltimore, MD"],"affiliations":[{"raw_affiliation_string":"Center for Language and Speech Processing, Johns Hopkins University, Baltimore, MD","institution_ids":["https://openalex.org/I145311948"]},{"raw_affiliation_string":"Human Language Technology Center of Excellence, Johns Hopkins University, Baltimore, MD","institution_ids":["https://openalex.org/I145311948"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5057010207","display_name":"Jes\u00fas Villalba","orcid":"https://orcid.org/0000-0001-9459-8426"},"institutions":[{"id":"https://openalex.org/I145311948","display_name":"Johns Hopkins University","ror":"https://ror.org/00za53h95","country_code":"US","type":"education","lineage":["https://openalex.org/I145311948"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jesus Villalba","raw_affiliation_strings":["Center for Language and Speech Processing, Johns Hopkins University, Baltimore, MD","Human Language Technology Center of Excellence, Johns Hopkins University, Baltimore, MD"],"affiliations":[{"raw_affiliation_string":"Center for Language and Speech Processing, Johns Hopkins University, Baltimore, MD","institution_ids":["https://openalex.org/I145311948"]},{"raw_affiliation_string":"Human Language Technology Center of Excellence, Johns Hopkins University, Baltimore, MD","institution_ids":["https://openalex.org/I145311948"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5050632169","display_name":"Najim Dehak","orcid":"https://orcid.org/0000-0002-4489-5753"},"institutions":[{"id":"https://openalex.org/I145311948","display_name":"Johns Hopkins University","ror":"https://ror.org/00za53h95","country_code":"US","type":"education","lineage":["https://openalex.org/I145311948"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Najim Dehak","raw_affiliation_strings":["Center for Language and Speech Processing, Johns Hopkins University, Baltimore, MD","Human Language Technology Center of Excellence, Johns Hopkins University, Baltimore, MD"],"affiliations":[{"raw_affiliation_string":"Center for Language and Speech Processing, Johns Hopkins University, Baltimore, MD","institution_ids":["https://openalex.org/I145311948"]},{"raw_affiliation_string":"Human Language Technology Center of Excellence, Johns Hopkins University, Baltimore, MD","institution_ids":["https://openalex.org/I145311948"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5071702566"],"corresponding_institution_ids":["https://openalex.org/I145311948"],"apc_list":null,"apc_paid":null,"fwci":0.4079,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.67468903,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":96},"biblio":{"volume":"12","issue":null,"first_page":"5994","last_page":"5998"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9998999834060669,"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":0.9998999834060669,"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/T10860","display_name":"Speech and Audio Processing","score":0.992900013923645,"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/T10028","display_name":"Topic Modeling","score":0.9927999973297119,"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":0.7728084325790405},{"id":"https://openalex.org/keywords/regularization","display_name":"Regularization (linguistics)","score":0.7260207533836365},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.7197842597961426},{"id":"https://openalex.org/keywords/focus","display_name":"Focus (optics)","score":0.6253321766853333},{"id":"https://openalex.org/keywords/connectionism","display_name":"Connectionism","score":0.6160988807678223},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.6084804534912109},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5796525478363037},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.45673584938049316},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.12620463967323303},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.0741262435913086}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7728084325790405},{"id":"https://openalex.org/C2776135515","wikidata":"https://www.wikidata.org/wiki/Q17143721","display_name":"Regularization (linguistics)","level":2,"score":0.7260207533836365},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.7197842597961426},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.6253321766853333},{"id":"https://openalex.org/C8521452","wikidata":"https://www.wikidata.org/wiki/Q203790","display_name":"Connectionism","level":3,"score":0.6160988807678223},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.6084804534912109},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5796525478363037},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.45673584938049316},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.12620463967323303},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.0741262435913086},{"id":"https://openalex.org/C120665830","wikidata":"https://www.wikidata.org/wiki/Q14620","display_name":"Optics","level":1,"score":0.0},{"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/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp39728.2021.9414648","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp39728.2021.9414648","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.46000000834465027,"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":41,"referenced_works":["https://openalex.org/W46679369","https://openalex.org/W854541894","https://openalex.org/W1494198834","https://openalex.org/W1787224781","https://openalex.org/W1915485278","https://openalex.org/W1932198206","https://openalex.org/W2102605133","https://openalex.org/W2133564696","https://openalex.org/W2172097686","https://openalex.org/W2251321385","https://openalex.org/W2327501763","https://openalex.org/W2739883972","https://openalex.org/W2741040846","https://openalex.org/W2799051177","https://openalex.org/W2892009249","https://openalex.org/W2897507397","https://openalex.org/W2936774411","https://openalex.org/W2944255943","https://openalex.org/W2962780374","https://openalex.org/W2962851944","https://openalex.org/W2963403868","https://openalex.org/W2963674932","https://openalex.org/W2964159778","https://openalex.org/W2964308564","https://openalex.org/W2972498556","https://openalex.org/W2972808286","https://openalex.org/W2972818416","https://openalex.org/W3007328579","https://openalex.org/W3015412285","https://openalex.org/W3047517563","https://openalex.org/W3103005696","https://openalex.org/W3127151792","https://openalex.org/W4385245566","https://openalex.org/W6601894380","https://openalex.org/W6623517193","https://openalex.org/W6679434410","https://openalex.org/W6685133223","https://openalex.org/W6691770337","https://openalex.org/W6739901393","https://openalex.org/W6754299077","https://openalex.org/W6781636266"],"related_works":["https://openalex.org/W4205841273","https://openalex.org/W4205525690","https://openalex.org/W1732468982","https://openalex.org/W1761388607","https://openalex.org/W1997922073","https://openalex.org/W350032239","https://openalex.org/W2604685715","https://openalex.org/W2412160900","https://openalex.org/W2136453575","https://openalex.org/W1596913645"],"abstract_inverted_index":{"This":[0],"paper":[1],"introduces":[2],"a":[3,87],"novel":[4],"method":[5,27,90],"to":[6,56,74,96,106,120],"diagnose":[7],"the":[8,31,42,48,51,58,62,68,80,85,103,107,113,126],"source-target":[9,37,69,98],"attention":[10,24,63,70,99],"in":[11],"state-of-the-art":[12],"end-to-end":[13],"speech":[14],"recognition":[15],"models":[16],"with":[17,125],"joint":[18],"connectionist":[19],"temporal":[20],"classification":[21],"(CTC)":[22],"and":[23,36,122,132],"training.":[25],"Our":[26],"is":[28,54,91],"based":[29],"on":[30,41,102,129],"fact":[32],"that":[33,67],"both,":[34],"CTC":[35,53,95],"attention,":[38,52],"are":[39,72],"acting":[40],"same":[43],"encoder":[44],"representations.":[45],"To":[46],"understand":[47],"functionality":[49],"of":[50,79],"applied":[55],"compute":[57],"token":[59,109],"posteriors":[60],"given":[61],"outputs.":[64],"We":[65],"found":[66],"heads":[71],"able":[73],"predict":[75],"several":[76],"tokens":[77],"ahead":[78],"current":[81],"one.":[82],"Inspired":[83],"by":[84,112],"observation,":[86],"new":[88],"regularization":[89,128],"proposed":[92,127],"which":[93],"leverages":[94],"make":[97],"more":[100],"focused":[101],"frames":[104],"corresponding":[105],"output":[108],"being":[110],"predicted":[111],"decoder.":[114],"Experiments":[115],"reveal":[116],"stable":[117],"improvements":[118],"up":[119],"7%":[121],"13%":[123],"relatively":[124],"TED-LIUM":[130],"2":[131],"Librispeech.":[133]},"counts_by_year":[{"year":2023,"cited_by_count":2},{"year":2021,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}