{"id":"https://openalex.org/W4372266510","doi":"https://doi.org/10.1109/icassp49357.2023.10095860","title":"Adaptable End-to-End ASR Models Using Replaceable Internal LMs and Residual Softmax","display_name":"Adaptable End-to-End ASR Models Using Replaceable Internal LMs and Residual Softmax","publication_year":2023,"publication_date":"2023-05-05","ids":{"openalex":"https://openalex.org/W4372266510","doi":"https://doi.org/10.1109/icassp49357.2023.10095860"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49357.2023.10095860","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49357.2023.10095860","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","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/A5068367173","display_name":"Keqi Deng","orcid":"https://orcid.org/0000-0003-1490-963X"},"institutions":[{"id":"https://openalex.org/I241749","display_name":"University of Cambridge","ror":"https://ror.org/013meh722","country_code":"GB","type":"education","lineage":["https://openalex.org/I241749"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Keqi Deng","raw_affiliation_strings":["University of Cambridge,Department of Engineering,Cambridge,UK","Department of Engineering, University of Cambridge, Cambridge, UK"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Cambridge,Department of Engineering,Cambridge,UK","institution_ids":["https://openalex.org/I241749"]},{"raw_affiliation_string":"Department of Engineering, University of Cambridge, Cambridge, UK","institution_ids":["https://openalex.org/I241749"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5002191410","display_name":"Philip C. Woodland","orcid":"https://orcid.org/0000-0001-9069-0225"},"institutions":[{"id":"https://openalex.org/I241749","display_name":"University of Cambridge","ror":"https://ror.org/013meh722","country_code":"GB","type":"education","lineage":["https://openalex.org/I241749"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Philip C. Woodland","raw_affiliation_strings":["University of Cambridge,Department of Engineering,Cambridge,UK","Department of Engineering, University of Cambridge, Cambridge, UK"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Cambridge,Department of Engineering,Cambridge,UK","institution_ids":["https://openalex.org/I241749"]},{"raw_affiliation_string":"Department of Engineering, University of Cambridge, Cambridge, UK","institution_ids":["https://openalex.org/I241749"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I241749"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":9,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"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/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/T11309","display_name":"Music and Audio Processing","score":0.9968000054359436,"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/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"}}],"keywords":[{"id":"https://openalex.org/keywords/softmax-function","display_name":"Softmax function","score":0.8808515071868896},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7566883563995361},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.6378991603851318},{"id":"https://openalex.org/keywords/residual","display_name":"Residual","score":0.6134674549102783},{"id":"https://openalex.org/keywords/end-to-end-principle","display_name":"End-to-end principle","score":0.5774937272071838},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5408099293708801},{"id":"https://openalex.org/keywords/domain-adaptation","display_name":"Domain adaptation","score":0.5352761149406433},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5205712914466858},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.5047086477279663},{"id":"https://openalex.org/keywords/domain","display_name":"Domain (mathematical analysis)","score":0.4897470474243164},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4634219706058502},{"id":"https://openalex.org/keywords/reduction","display_name":"Reduction (mathematics)","score":0.43564850091934204},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.11505323648452759},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.0996970534324646},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.07142385840415955}],"concepts":[{"id":"https://openalex.org/C188441871","wikidata":"https://www.wikidata.org/wiki/Q7554146","display_name":"Softmax function","level":3,"score":0.8808515071868896},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7566883563995361},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.6378991603851318},{"id":"https://openalex.org/C155512373","wikidata":"https://www.wikidata.org/wiki/Q287450","display_name":"Residual","level":2,"score":0.6134674549102783},{"id":"https://openalex.org/C74296488","wikidata":"https://www.wikidata.org/wiki/Q2527392","display_name":"End-to-end principle","level":2,"score":0.5774937272071838},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5408099293708801},{"id":"https://openalex.org/C2776434776","wikidata":"https://www.wikidata.org/wiki/Q19246213","display_name":"Domain adaptation","level":3,"score":0.5352761149406433},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5205712914466858},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.5047086477279663},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.4897470474243164},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4634219706058502},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.43564850091934204},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.11505323648452759},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.0996970534324646},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.07142385840415955},{"id":"https://openalex.org/C95623464","wikidata":"https://www.wikidata.org/wiki/Q1096149","display_name":"Classifier (UML)","level":2,"score":0.0},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp49357.2023.10095860","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49357.2023.10095860","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Quality Education","score":0.6800000071525574,"id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[{"id":"https://openalex.org/F4320320273","display_name":"University of Cambridge","ror":"https://ror.org/013meh722"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":36,"referenced_works":["https://openalex.org/W46679369","https://openalex.org/W854541894","https://openalex.org/W1494198834","https://openalex.org/W1915251500","https://openalex.org/W2016589492","https://openalex.org/W2087006792","https://openalex.org/W2143612262","https://openalex.org/W2166637769","https://openalex.org/W2194775991","https://openalex.org/W2766219058","https://openalex.org/W2888779557","https://openalex.org/W2933138175","https://openalex.org/W2962780374","https://openalex.org/W2963240019","https://openalex.org/W2972847736","https://openalex.org/W3008037978","https://openalex.org/W3036601975","https://openalex.org/W3044057088","https://openalex.org/W3097777922","https://openalex.org/W3152221657","https://openalex.org/W3160475509","https://openalex.org/W3163560333","https://openalex.org/W3198439131","https://openalex.org/W3198771897","https://openalex.org/W3205201903","https://openalex.org/W3205788551","https://openalex.org/W4205442444","https://openalex.org/W4210758944","https://openalex.org/W4226158189","https://openalex.org/W4283031490","https://openalex.org/W4385245566","https://openalex.org/W6601894380","https://openalex.org/W6623517193","https://openalex.org/W6739901393","https://openalex.org/W6780218876","https://openalex.org/W6780957418"],"related_works":["https://openalex.org/W3107204728","https://openalex.org/W4287591324","https://openalex.org/W3108503355","https://openalex.org/W4226420367","https://openalex.org/W2962876041","https://openalex.org/W2898268032","https://openalex.org/W1773599773","https://openalex.org/W2268150819","https://openalex.org/W2916997151","https://openalex.org/W2949174760"],"abstract_inverted_index":{"End-to-end":[0],"(E2E)":[1],"automatic":[2],"speech":[3],"recognition":[4],"(ASR)":[5],"implicitly":[6],"learns":[7],"the":[8,55,69,97,110,116,126,135],"token":[9],"sequence":[10],"distribution":[11],"of":[12,60],"paired":[13],"audio-transcript":[14],"training":[15,25],"data.":[16],"However,":[17],"it":[18,50],"still":[19,32],"suffers":[20],"from":[21,24],"domain":[22,29,74,99],"shifts":[23],"to":[26,52,94,96],"testing,":[27],"and":[28,129],"adaptation":[30],"is":[31,76,87],"challenging.":[33],"To":[34],"alleviate":[35],"this":[36,38,79],"problem,":[37],"paper":[39,80],"designs":[40],"a":[41,65,73,82,120,130],"replaceable":[42],"internal":[43,56],"language":[44,57],"model":[45,58],"(RILM)":[46],"method,":[47],"which":[48],"makes":[49],"feasible":[51],"directly":[53],"replace":[54],"(LM)":[59],"E2E":[61,91,105],"ASR":[62,92,106,141],"models":[63,93,107],"with":[64],"target-domain":[66],"LM":[67],"in":[68],"decoding":[70],"stage":[71],"when":[72],"shift":[75],"encountered.":[77],"Furthermore,":[78],"proposes":[81],"residual":[83],"softmax":[84],"(R-softmax)":[85],"that":[86,115],"designed":[88],"for":[89],"CTC-based":[90],"adapt":[95],"target":[98],"without":[100],"re-training":[101],"during":[102],"inference.":[103],"For":[104],"trained":[108],"on":[109,125,134],"LibriSpeech":[111],"corpus,":[112],"experiments":[113],"showed":[114],"proposed":[117],"methods":[118],"gave":[119],"2.6%":[121],"absolute":[122],"WER":[123,132],"reduction":[124,133],"Switchboard":[127],"data":[128],"1.0%":[131],"AESRC2020":[136],"corpus":[137],"while":[138],"maintaining":[139],"intra-domain":[140],"results.":[142]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":5},{"year":2023,"cited_by_count":2}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}