{"id":"https://openalex.org/W4319862233","doi":"https://doi.org/10.1109/slt54892.2023.10023442","title":"Transformer-Based Lip-Reading with Regularized Dropout and Relaxed Attention","display_name":"Transformer-Based Lip-Reading with Regularized Dropout and Relaxed Attention","publication_year":2023,"publication_date":"2023-01-09","ids":{"openalex":"https://openalex.org/W4319862233","doi":"https://doi.org/10.1109/slt54892.2023.10023442"},"language":"en","primary_location":{"id":"doi:10.1109/slt54892.2023.10023442","is_oa":false,"landing_page_url":"https://doi.org/10.1109/slt54892.2023.10023442","pdf_url":null,"source":{"id":"https://openalex.org/S4363605953","display_name":"2022 IEEE Spoken Language Technology Workshop (SLT)","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":"2022 IEEE Spoken Language Technology Workshop (SLT)","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/A5100736672","display_name":"Zhengyang Li","orcid":"https://orcid.org/0009-0007-6764-933X"},"institutions":[{"id":"https://openalex.org/I94509681","display_name":"Technische Universit\u00e4t Braunschweig","ror":"https://ror.org/010nsgg66","country_code":"DE","type":"education","lineage":["https://openalex.org/I94509681"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Zhengyang Li","raw_affiliation_strings":["Technische Universität Braunschweig, Institute for Communications Technology,Braunschweig,Germany,38106"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Technische Universität Braunschweig, Institute for Communications Technology,Braunschweig,Germany,38106","institution_ids":["https://openalex.org/I94509681"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5039030019","display_name":"Timo Lohrenz","orcid":"https://orcid.org/0000-0001-7103-4302"},"institutions":[{"id":"https://openalex.org/I94509681","display_name":"Technische Universit\u00e4t Braunschweig","ror":"https://ror.org/010nsgg66","country_code":"DE","type":"education","lineage":["https://openalex.org/I94509681"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Timo Lohrenz","raw_affiliation_strings":["Technische Universität Braunschweig, Institute for Communications Technology,Braunschweig,Germany,38106"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Technische Universität Braunschweig, Institute for Communications Technology,Braunschweig,Germany,38106","institution_ids":["https://openalex.org/I94509681"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5052937699","display_name":"Matthias Dunkelberg","orcid":null},"institutions":[{"id":"https://openalex.org/I94509681","display_name":"Technische Universit\u00e4t Braunschweig","ror":"https://ror.org/010nsgg66","country_code":"DE","type":"education","lineage":["https://openalex.org/I94509681"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Matthias Dunkelberg","raw_affiliation_strings":["Technische Universität Braunschweig, Institute for Communications Technology,Braunschweig,Germany,38106"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Technische Universität Braunschweig, Institute for Communications Technology,Braunschweig,Germany,38106","institution_ids":["https://openalex.org/I94509681"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5002593702","display_name":"Tim Fingscheidt","orcid":"https://orcid.org/0000-0002-8895-5041"},"institutions":[{"id":"https://openalex.org/I94509681","display_name":"Technische Universit\u00e4t Braunschweig","ror":"https://ror.org/010nsgg66","country_code":"DE","type":"education","lineage":["https://openalex.org/I94509681"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Tim Fingscheidt","raw_affiliation_strings":["Technische Universität Braunschweig, Institute for Communications Technology,Braunschweig,Germany,38106"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Technische Universität Braunschweig, Institute for Communications Technology,Braunschweig,Germany,38106","institution_ids":["https://openalex.org/I94509681"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I94509681"],"apc_list":null,"apc_paid":null,"fwci":1.45,"has_fulltext":false,"cited_by_count":6,"citation_normalized_percentile":{"value":0.83657451,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"723","last_page":"730"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10860","display_name":"Speech and Audio Processing","score":1.0,"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":1.0,"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":0.9991000294685364,"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/T11309","display_name":"Music and Audio Processing","score":0.995199978351593,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6783298254013062},{"id":"https://openalex.org/keywords/regularization","display_name":"Regularization (linguistics)","score":0.6006077527999878},{"id":"https://openalex.org/keywords/smoothing","display_name":"Smoothing","score":0.5640769004821777},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.517859935760498},{"id":"https://openalex.org/keywords/decoding-methods","display_name":"Decoding methods","score":0.46987730264663696},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4665849804878235},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.46610817313194275},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.4236270785331726},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.37426167726516724},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.3367984890937805},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.21510690450668335},{"id":"https://openalex.org/keywords/computer-vision","display_name":"Computer vision","score":0.15916121006011963}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6783298254013062},{"id":"https://openalex.org/C2776135515","wikidata":"https://www.wikidata.org/wiki/Q17143721","display_name":"Regularization (linguistics)","level":2,"score":0.6006077527999878},{"id":"https://openalex.org/C3770464","wikidata":"https://www.wikidata.org/wiki/Q775963","display_name":"Smoothing","level":2,"score":0.5640769004821777},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.517859935760498},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.46987730264663696},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4665849804878235},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.46610817313194275},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.4236270785331726},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.37426167726516724},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3367984890937805},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.21510690450668335},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.15916121006011963},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"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/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/slt54892.2023.10023442","is_oa":false,"landing_page_url":"https://doi.org/10.1109/slt54892.2023.10023442","pdf_url":null,"source":{"id":"https://openalex.org/S4363605953","display_name":"2022 IEEE Spoken Language Technology Workshop (SLT)","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":"2022 IEEE Spoken Language Technology Workshop (SLT)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4","score":0.8899999856948853}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":57,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W1915251500","https://openalex.org/W2095705004","https://openalex.org/W2526425061","https://openalex.org/W2551572271","https://openalex.org/W2577366047","https://openalex.org/W2808631503","https://openalex.org/W2888779557","https://openalex.org/W2890952074","https://openalex.org/W2891205112","https://openalex.org/W2892009249","https://openalex.org/W2896457183","https://openalex.org/W2912984882","https://openalex.org/W2948210185","https://openalex.org/W2963250244","https://openalex.org/W2963362078","https://openalex.org/W2963528589","https://openalex.org/W2963654130","https://openalex.org/W2963654155","https://openalex.org/W2975381464","https://openalex.org/W3006974783","https://openalex.org/W3008037978","https://openalex.org/W3015830103","https://openalex.org/W3016011581","https://openalex.org/W3035003500","https://openalex.org/W3096318498","https://openalex.org/W3097777922","https://openalex.org/W3097973766","https://openalex.org/W3123318516","https://openalex.org/W3162293946","https://openalex.org/W3163169798","https://openalex.org/W3174864715","https://openalex.org/W3197567540","https://openalex.org/W3198871389","https://openalex.org/W4221153068","https://openalex.org/W4226370720","https://openalex.org/W4297798436","https://openalex.org/W4307286264","https://openalex.org/W4312638101","https://openalex.org/W4385245566","https://openalex.org/W6674330103","https://openalex.org/W6727785063","https://openalex.org/W6732696085","https://openalex.org/W6734491695","https://openalex.org/W6739901393","https://openalex.org/W6750169759","https://openalex.org/W6754392867","https://openalex.org/W6754420807","https://openalex.org/W6755207826","https://openalex.org/W6763342845","https://openalex.org/W6764043288","https://openalex.org/W6768080748","https://openalex.org/W6777970034","https://openalex.org/W6779101013","https://openalex.org/W6797110910","https://openalex.org/W6798017321","https://openalex.org/W6810168380"],"related_works":["https://openalex.org/W1978572805","https://openalex.org/W2383807498","https://openalex.org/W1997992934","https://openalex.org/W1987225439","https://openalex.org/W4238188170","https://openalex.org/W2125114371","https://openalex.org/W2019977573","https://openalex.org/W2149980199","https://openalex.org/W4213275102","https://openalex.org/W2151138761"],"abstract_inverted_index":{"End-to-end":[0],"automatic":[1],"lip-reading":[2,37],"usually":[3],"comprises":[4],"an":[5,9,84],"encoder-decoder":[6],"model":[7,57],"and":[8,90],"optional":[10],"external":[11,55],"language":[12,56],"model.":[13],"In":[14,80],"this":[15],"work,":[16],"we":[17,28,102,125],"introduce":[18],"two":[19,67],"regularization":[20],"methods":[21],"to":[22,35,38,63],"the":[23,30,44,61,77,92,108,140],"field":[24],"of":[25,107,115,136],"lip-reading:":[26],"First,":[27],"apply":[29],"regularized":[31],"dropout":[32],"(R-Drop)":[33],"method":[34],"transformer-based":[36],"improve":[39],"their":[40],"training-inference":[41],"consistency.":[42],"Second,":[43],"relaxed":[45],"attention":[46,93,98],"technique":[47],"is":[48,150],"applied":[49],"during":[50,99],"training":[51,100,137],"for":[52],"a":[53,104,111],"better":[54],"integration.":[58],"We":[59],"are":[60,126],"first":[62],"show":[64],"that":[65],"these":[66],"complementary":[68],"approaches":[69],"yield":[70],"particu1arly":[71],"strong":[72],"performance":[73],"if":[74],"combined":[75],"in":[76,95],"right":[78],"manner.":[79],"particular,":[81],"by":[82],"adding":[83],"additional":[85],"R":[86],"-":[87],"Drop":[88],"loss":[89],"smoothing":[91],"weights":[94],"cross":[96],"multi-head":[97],"only,":[101],"achieve":[103],"new":[105],"state":[106],"art":[109],"with":[110,129],"word":[112],"error":[113],"rate":[114],"22.2%":[116],"on":[117],"Lip":[118],"Reading":[119],"Sentences":[120],"2":[121],"(LRS2).":[122],"On":[123],"LRS3,":[124],"2nd":[127],"ranked":[128],"25.5%":[130],"WER":[131],"using":[132],"only":[133],"1,759":[134],"h":[135],"data,":[138],"while":[139],"1":[141],"st":[142],"rank":[143],"uses":[144],"about":[145],"90,000":[146],"h.":[147],"Our":[148],"code":[149],"available":[151],"at":[152],"GitHub.11https://github.com/ifnspaml/Lipreading-RDrop-RA":[153]},"counts_by_year":[{"year":2025,"cited_by_count":4},{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":1}],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}