{"id":"https://openalex.org/W2714176837","doi":"https://doi.org/10.21437/interspeech.2017-103","title":"Improved Subword Modeling for WFST-Based Speech Recognition","display_name":"Improved Subword Modeling for WFST-Based Speech Recognition","publication_year":2017,"publication_date":"2017-08-16","ids":{"openalex":"https://openalex.org/W2714176837","doi":"https://doi.org/10.21437/interspeech.2017-103","mag":"2714176837"},"language":"en","primary_location":{"id":"doi:10.21437/interspeech.2017-103","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2017-103","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2017","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://research.aalto.fi/en/publications/ed43f22c-f5bd-45ad-99a7-628f82f2283c","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5074707861","display_name":"Peter Smit","orcid":"https://orcid.org/0000-0001-7611-1477"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Peter Smit","raw_affiliation_strings":["Department of Signal Processing and Acoustics"],"affiliations":[{"raw_affiliation_string":"Department of Signal Processing and Acoustics","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5003986900","display_name":"S\u00e1mi Virpioja","orcid":"https://orcid.org/0000-0002-3568-150X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sami Virpioja","raw_affiliation_strings":["Department of Signal Processing and Acoustics"],"affiliations":[{"raw_affiliation_string":"Department of Signal Processing and Acoustics","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5043424064","display_name":"Mikko Kurimo","orcid":"https://orcid.org/0000-0001-5278-7974"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mikko Kurimo","raw_affiliation_strings":["Department of Signal Processing and Acoustics"],"affiliations":[{"raw_affiliation_string":"Department of Signal Processing and Acoustics","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5074707861"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":6.4433,"has_fulltext":false,"cited_by_count":42,"citation_normalized_percentile":{"value":0.97150622,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":94,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"2551","last_page":"2555"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9599999785423279,"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":0.9599999785423279,"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":0.7306923866271973},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.515403151512146},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.47783422470092773},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.4338380992412567},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.3741312623023987}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7306923866271973},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.515403151512146},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.47783422470092773},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4338380992412567},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3741312623023987}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.21437/interspeech.2017-103","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2017-103","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2017","raw_type":"proceedings-article"},{"id":"pmh:oai:aaltodoc.aalto.fi:123456789/28342","is_oa":true,"landing_page_url":"https://research.aalto.fi/en/publications/ed43f22c-f5bd-45ad-99a7-628f82f2283c","pdf_url":null,"source":{"id":"https://openalex.org/S4306401662","display_name":"Aaltodoc (Aalto University)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I9927081","host_organization_name":"Aalto University","host_organization_lineage":["https://openalex.org/I9927081"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"publishedVersion"}],"best_oa_location":{"id":"pmh:oai:aaltodoc.aalto.fi:123456789/28342","is_oa":true,"landing_page_url":"https://research.aalto.fi/en/publications/ed43f22c-f5bd-45ad-99a7-628f82f2283c","pdf_url":null,"source":{"id":"https://openalex.org/S4306401662","display_name":"Aaltodoc (Aalto University)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I9927081","host_organization_name":"Aalto University","host_organization_lineage":["https://openalex.org/I9927081"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"publishedVersion"},"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320321394","display_name":"Aalto-Yliopisto","ror":"https://ror.org/020hwjq30"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":23,"referenced_works":["https://openalex.org/W46679369","https://openalex.org/W309335912","https://openalex.org/W1425721496","https://openalex.org/W1524333225","https://openalex.org/W1895481600","https://openalex.org/W1987408080","https://openalex.org/W2008160512","https://openalex.org/W2032942114","https://openalex.org/W2042783153","https://openalex.org/W2053306448","https://openalex.org/W2070737455","https://openalex.org/W2117621558","https://openalex.org/W2142937603","https://openalex.org/W2152753389","https://openalex.org/W2156700117","https://openalex.org/W2251903896","https://openalex.org/W2345764277","https://openalex.org/W2401896149","https://openalex.org/W2402261002","https://openalex.org/W2514741789","https://openalex.org/W2522481886","https://openalex.org/W2573287566","https://openalex.org/W2962784628"],"related_works":["https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W2358668433","https://openalex.org/W2376932109","https://openalex.org/W2382290278","https://openalex.org/W2350741829","https://openalex.org/W2033914206","https://openalex.org/W2042327336","https://openalex.org/W3192589309"],"abstract_inverted_index":{"Because":[0],"in":[1,18,51,146],"agglutinative":[2],"languages":[3],"the":[4,22,42,52,64,79,87,130,171,175],"number":[5],"of":[6,25,31,41,73,92,125],"observed":[7],"word":[8,74,105],"forms":[9],"is":[10,98,148,180],"very":[11],"high,":[12],"subword":[13,26,49,65,94,117,132,140],"units":[14,27,66],"are":[15,162],"often":[16],"utilized":[17],"speech":[19,158],"recognition.":[20],"However,":[21],"proper":[23],"use":[24],"requires":[28],"careful":[29],"consideration":[30],"details":[32],"such":[33],"as":[34],"silence":[35],"modeling,":[36],"position-dependent":[37],"phones,":[38],"and":[39,77,127,138,174],"combination":[40],"units.":[43],"In":[44,119],"this":[45],"paper,":[46],"we":[47],"implement":[48],"modeling":[50],"Kaldi":[53],"toolkit":[54],"by":[55,59,82,103],"creating":[56],"modified":[57],"lexicon":[58],"finite-state":[60],"transducers":[61],"to":[62,86],"represent":[63],"correctly.":[67],"We":[68,108],"experiment":[69],"with":[70],"multiple":[71],"types":[72],"boundary":[75],"markers":[76],"achieve":[78],"best":[80,131],"results":[81,161],"adding":[83],"a":[84,93,104,123,149,155],"marker":[85],"left":[88],"or":[89,101],"right":[90],"side":[91],"unit":[95],"whenever":[96],"it":[97],"not":[99],"preceded":[100],"followed":[102],"boundary,":[106],"respectively.":[107],"also":[109,163],"compare":[110],"three":[111],"different":[112],"toolkits":[113],"that":[114],"provide":[115],"data-driven":[116],"segmentations.":[118],"our":[120],"experiments":[121],"on":[122,177],"variety":[124],"Finnish":[126,156],"Estonian":[128],"datasets,":[129,173],"models":[133,137,153],"do":[134],"outperform":[135],"word-based":[136,152],"naive":[139],"implementations.":[141],"The":[142,160],"largest":[143],"relative":[144],"reduction":[145],"WER":[147],"23%":[150],"over":[151],"for":[154,170],"read":[157],"dataset.":[159],"better":[164],"than":[165,182],"any":[166],"previously":[167],"published":[168],"ones":[169],"same":[172],"improvement":[176],"all":[178],"datasets":[179],"more":[181],"5%.":[183]},"counts_by_year":[{"year":2025,"cited_by_count":2},{"year":2023,"cited_by_count":2},{"year":2022,"cited_by_count":4},{"year":2021,"cited_by_count":3},{"year":2020,"cited_by_count":11},{"year":2019,"cited_by_count":12},{"year":2018,"cited_by_count":4},{"year":2017,"cited_by_count":4}],"updated_date":"2026-04-04T06:10:10.580331","created_date":"2025-10-10T00:00:00"}