{"id":"https://openalex.org/W4319862718","doi":"https://doi.org/10.1109/slt54892.2023.10023257","title":"Four-in-One: a Joint Approach to Inverse Text Normalization, Punctuation, Capitalization, and Disfluency for Automatic Speech Recognition","display_name":"Four-in-One: a Joint Approach to Inverse Text Normalization, Punctuation, Capitalization, and Disfluency for Automatic Speech Recognition","publication_year":2023,"publication_date":"2023-01-09","ids":{"openalex":"https://openalex.org/W4319862718","doi":"https://doi.org/10.1109/slt54892.2023.10023257"},"language":"en","primary_location":{"id":"doi:10.1109/slt54892.2023.10023257","is_oa":false,"landing_page_url":"https://doi.org/10.1109/slt54892.2023.10023257","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":"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/A5088618982","display_name":"Sharman Tan","orcid":null},"institutions":[{"id":"https://openalex.org/I4210105678","display_name":"Microsoft (Finland)","ror":"https://ror.org/01nehjf29","country_code":"FI","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210105678"]}],"countries":["FI"],"is_corresponding":true,"raw_author_name":"Sharman Tan","raw_affiliation_strings":["Microsoft Corporation"],"affiliations":[{"raw_affiliation_string":"Microsoft Corporation","institution_ids":["https://openalex.org/I4210105678"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5044402723","display_name":"Piyush Behre","orcid":null},"institutions":[{"id":"https://openalex.org/I4210105678","display_name":"Microsoft (Finland)","ror":"https://ror.org/01nehjf29","country_code":"FI","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210105678"]}],"countries":["FI"],"is_corresponding":false,"raw_author_name":"Piyush Behre","raw_affiliation_strings":["Microsoft Corporation"],"affiliations":[{"raw_affiliation_string":"Microsoft Corporation","institution_ids":["https://openalex.org/I4210105678"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5033810208","display_name":"Nick Kibre","orcid":null},"institutions":[{"id":"https://openalex.org/I4210105678","display_name":"Microsoft (Finland)","ror":"https://ror.org/01nehjf29","country_code":"FI","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210105678"]}],"countries":["FI"],"is_corresponding":false,"raw_author_name":"Nick Kibre","raw_affiliation_strings":["Microsoft Corporation"],"affiliations":[{"raw_affiliation_string":"Microsoft Corporation","institution_ids":["https://openalex.org/I4210105678"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5026044527","display_name":"Issac Alphonso","orcid":null},"institutions":[{"id":"https://openalex.org/I4210105678","display_name":"Microsoft (Finland)","ror":"https://ror.org/01nehjf29","country_code":"FI","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210105678"]}],"countries":["FI"],"is_corresponding":false,"raw_author_name":"Issac Alphonso","raw_affiliation_strings":["Microsoft Corporation"],"affiliations":[{"raw_affiliation_string":"Microsoft Corporation","institution_ids":["https://openalex.org/I4210105678"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5113757183","display_name":"Shuangyu Chang","orcid":null},"institutions":[{"id":"https://openalex.org/I4210105678","display_name":"Microsoft (Finland)","ror":"https://ror.org/01nehjf29","country_code":"FI","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210105678"]}],"countries":["FI"],"is_corresponding":false,"raw_author_name":"Shuangyu Chang","raw_affiliation_strings":["Microsoft Corporation"],"affiliations":[{"raw_affiliation_string":"Microsoft Corporation","institution_ids":["https://openalex.org/I4210105678"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5088618982"],"corresponding_institution_ids":["https://openalex.org/I4210105678"],"apc_list":null,"apc_paid":null,"fwci":0.5643,"has_fulltext":false,"cited_by_count":5,"citation_normalized_percentile":{"value":0.59812598,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"677","last_page":"684"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","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/T10028","display_name":"Topic Modeling","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/T10181","display_name":"Natural Language Processing Techniques","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/T12031","display_name":"Speech and dialogue systems","score":0.9997000098228455,"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":0.8431011438369751},{"id":"https://openalex.org/keywords/punctuation","display_name":"Punctuation","score":0.8249884247779846},{"id":"https://openalex.org/keywords/disk-formatting","display_name":"Disk formatting","score":0.7423419952392578},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.706545889377594},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5622297525405884},{"id":"https://openalex.org/keywords/normalization","display_name":"Normalization (sociology)","score":0.5102524757385254},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.5003693103790283},{"id":"https://openalex.org/keywords/plain-text","display_name":"Plain text","score":0.45931532979011536},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.4536544382572174},{"id":"https://openalex.org/keywords/spoken-language","display_name":"Spoken language","score":0.41678160429000854},{"id":"https://openalex.org/keywords/lemmatisation","display_name":"Lemmatisation","score":0.4124409556388855}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8431011438369751},{"id":"https://openalex.org/C540372491","wikidata":"https://www.wikidata.org/wiki/Q82622","display_name":"Punctuation","level":2,"score":0.8249884247779846},{"id":"https://openalex.org/C88006597","wikidata":"https://www.wikidata.org/wiki/Q690117","display_name":"Disk formatting","level":2,"score":0.7423419952392578},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.706545889377594},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5622297525405884},{"id":"https://openalex.org/C136886441","wikidata":"https://www.wikidata.org/wiki/Q926129","display_name":"Normalization (sociology)","level":2,"score":0.5102524757385254},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.5003693103790283},{"id":"https://openalex.org/C46503548","wikidata":"https://www.wikidata.org/wiki/Q1145976","display_name":"Plain text","level":3,"score":0.45931532979011536},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.4536544382572174},{"id":"https://openalex.org/C2776230583","wikidata":"https://www.wikidata.org/wiki/Q1322198","display_name":"Spoken language","level":2,"score":0.41678160429000854},{"id":"https://openalex.org/C161831844","wikidata":"https://www.wikidata.org/wiki/Q2554325","display_name":"Lemmatisation","level":2,"score":0.4124409556388855},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C19165224","wikidata":"https://www.wikidata.org/wiki/Q23404","display_name":"Anthropology","level":1,"score":0.0},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.0},{"id":"https://openalex.org/C144024400","wikidata":"https://www.wikidata.org/wiki/Q21201","display_name":"Sociology","level":0,"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/C148730421","wikidata":"https://www.wikidata.org/wiki/Q141090","display_name":"Encryption","level":2,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/slt54892.2023.10023257","is_oa":false,"landing_page_url":"https://doi.org/10.1109/slt54892.2023.10023257","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":0.6399999856948853}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":47,"referenced_works":["https://openalex.org/W94670513","https://openalex.org/W97072897","https://openalex.org/W2101446498","https://openalex.org/W2103085228","https://openalex.org/W2112078429","https://openalex.org/W2117682636","https://openalex.org/W2127305722","https://openalex.org/W2128970689","https://openalex.org/W2135799088","https://openalex.org/W2139828426","https://openalex.org/W2153895584","https://openalex.org/W2161274063","https://openalex.org/W2166637769","https://openalex.org/W2169187092","https://openalex.org/W2419539795","https://openalex.org/W2531638282","https://openalex.org/W2613000335","https://openalex.org/W2620949368","https://openalex.org/W2750292073","https://openalex.org/W2767341768","https://openalex.org/W2945656493","https://openalex.org/W2949615363","https://openalex.org/W2963729456","https://openalex.org/W2963995225","https://openalex.org/W2973194065","https://openalex.org/W3016128928","https://openalex.org/W3034323214","https://openalex.org/W3045571807","https://openalex.org/W3094475872","https://openalex.org/W3094965760","https://openalex.org/W3160789530","https://openalex.org/W4205807230","https://openalex.org/W4221142855","https://openalex.org/W4226272674","https://openalex.org/W4292595872","https://openalex.org/W4295276571","https://openalex.org/W4385245566","https://openalex.org/W6603931906","https://openalex.org/W6605854414","https://openalex.org/W6632455782","https://openalex.org/W6684662674","https://openalex.org/W6717262007","https://openalex.org/W6729005282","https://openalex.org/W6731992975","https://openalex.org/W6732408451","https://openalex.org/W6753719773","https://openalex.org/W6755207826"],"related_works":["https://openalex.org/W2936002343","https://openalex.org/W2188883480","https://openalex.org/W1592364192","https://openalex.org/W656840002","https://openalex.org/W1605117403","https://openalex.org/W2381416480","https://openalex.org/W2380599343","https://openalex.org/W2181793145","https://openalex.org/W3034646373","https://openalex.org/W187699659"],"abstract_inverted_index":{"Features":[0],"such":[1],"as":[2],"punctuation,":[3,75],"capitalization,":[4,76],"and":[5,14,31,77,88],"formatting":[6,35],"of":[7,29],"entities":[8],"are":[9],"important":[10],"for":[11,70],"readability,":[12],"understanding,":[13],"natural":[15],"language":[16],"processing":[17],"tasks.":[18],"However,":[19],"Automatic":[20],"Speech":[21],"Recognition":[22],"(ASR)":[23],"systems":[24],"produce":[25,67],"spoken-form":[26],"text":[27,51,72,87],"devoid":[28],"formatting,":[30],"tagging":[32,63,110],"approaches":[33],"to":[34,65,84,96],"address":[36],"just":[37],"one":[38],"or":[39,113],"two":[40],"features":[41],"at":[42],"a":[43,54,60],"time.":[44],"In":[45],"this":[46],"paper,":[47],"we":[48,58,80],"unify":[49],"spoken-to-written":[50],"conversion":[52],"via":[53],"two-stage":[55],"process:":[56],"First,":[57],"use":[59,89],"single":[61],"transformer":[62],"model":[64],"jointly":[66],"token-level":[68],"tags":[69,83],"inverse":[71],"normalization":[73],"(ITN),":[74],"disfluencies.":[78],"Then,":[79],"apply":[81],"the":[82],"generate":[85],"written-form":[86],"weighted":[90],"finite":[91],"state":[92],"transducer":[93],"(WFST)":[94],"grammars":[95],"format":[97],"tagged":[98],"ITN":[99],"entity":[100],"spans.":[101],"Despite":[102],"joining":[103],"four":[104,119],"models":[105,116],"into":[106],"one,":[107],"our":[108],"unified":[109],"approach":[111],"matches":[112],"outperforms":[114],"task-specific":[115],"across":[117,125],"all":[118],"tasks":[120],"on":[121],"benchmark":[122],"test":[123],"sets":[124],"several":[126],"domains.":[127]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2023,"cited_by_count":4}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}