{"id":"https://openalex.org/W2508310510","doi":"https://doi.org/10.21437/interspeech.2016-1010","title":"Improving Under-Resourced Language ASR Through Latent Subword Unit Space Discovery","display_name":"Improving Under-Resourced Language ASR Through Latent Subword Unit Space Discovery","publication_year":2016,"publication_date":"2016-08-29","ids":{"openalex":"https://openalex.org/W2508310510","doi":"https://doi.org/10.21437/interspeech.2016-1010","mag":"2508310510"},"language":"en","primary_location":{"id":"doi:10.21437/interspeech.2016-1010","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2016-1010","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2016","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"http://infoscience.epfl.ch/record/219850","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5044443491","display_name":"Marzieh Razavi","orcid":"https://orcid.org/0000-0001-6332-9661"},"institutions":[{"id":"https://openalex.org/I5124864","display_name":"\u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne","ror":"https://ror.org/02s376052","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I5124864"]},{"id":"https://openalex.org/I7495430","display_name":"Idiap Research Institute","ror":"https://ror.org/05932h694","country_code":"CH","type":"facility","lineage":["https://openalex.org/I7495430"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Marzieh Razavi","raw_affiliation_strings":["Ecole Polytechnique F\u00e9d\u00e9rale de Lausanne (EPFL), CH-1015 Lausanne, Switzerland","Idiap Research Institute, CH-1920 Martigny, Switzerland"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Ecole Polytechnique F\u00e9d\u00e9rale de Lausanne (EPFL), CH-1015 Lausanne, Switzerland","institution_ids":["https://openalex.org/I5124864"]},{"raw_affiliation_string":"Idiap Research Institute, CH-1920 Martigny, Switzerland","institution_ids":["https://openalex.org/I7495430"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5043551083","display_name":"Mathew Magimai.-Doss","orcid":"https://orcid.org/0000-0002-8714-1409"},"institutions":[{"id":"https://openalex.org/I7495430","display_name":"Idiap Research Institute","ror":"https://ror.org/05932h694","country_code":"CH","type":"facility","lineage":["https://openalex.org/I7495430"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Mathew Magimai-Doss","raw_affiliation_strings":["Idiap Research Institute, CH-1920 Martigny, Switzerland"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Idiap Research Institute, CH-1920 Martigny, Switzerland","institution_ids":["https://openalex.org/I7495430"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.06369527,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"3873","last_page":"3877"},"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":0.9994000196456909,"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":0.9965999722480774,"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":0.8274667263031006},{"id":"https://openalex.org/keywords/phone","display_name":"Phone","score":0.7888695001602173},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.5880622863769531},{"id":"https://openalex.org/keywords/space","display_name":"Space (punctuation)","score":0.5605190992355347},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5551241040229797},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5219799280166626},{"id":"https://openalex.org/keywords/cluster-analysis","display_name":"Cluster analysis","score":0.5159485936164856},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.48223695158958435},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.480143278837204},{"id":"https://openalex.org/keywords/acoustic-model","display_name":"Acoustic model","score":0.47586148977279663},{"id":"https://openalex.org/keywords/hidden-markov-model","display_name":"Hidden Markov model","score":0.47060713171958923},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.445494681596756},{"id":"https://openalex.org/keywords/adaptation","display_name":"Adaptation (eye)","score":0.44384193420410156},{"id":"https://openalex.org/keywords/speech-processing","display_name":"Speech processing","score":0.2750924229621887},{"id":"https://openalex.org/keywords/linguistics","display_name":"Linguistics","score":0.16912785172462463}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8274667263031006},{"id":"https://openalex.org/C2778707766","wikidata":"https://www.wikidata.org/wiki/Q202064","display_name":"Phone","level":2,"score":0.7888695001602173},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5880622863769531},{"id":"https://openalex.org/C2778572836","wikidata":"https://www.wikidata.org/wiki/Q380933","display_name":"Space (punctuation)","level":2,"score":0.5605190992355347},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5551241040229797},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5219799280166626},{"id":"https://openalex.org/C73555534","wikidata":"https://www.wikidata.org/wiki/Q622825","display_name":"Cluster analysis","level":2,"score":0.5159485936164856},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.48223695158958435},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.480143278837204},{"id":"https://openalex.org/C155635449","wikidata":"https://www.wikidata.org/wiki/Q4674699","display_name":"Acoustic model","level":3,"score":0.47586148977279663},{"id":"https://openalex.org/C23224414","wikidata":"https://www.wikidata.org/wiki/Q176769","display_name":"Hidden Markov model","level":2,"score":0.47060713171958923},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.445494681596756},{"id":"https://openalex.org/C139807058","wikidata":"https://www.wikidata.org/wiki/Q352374","display_name":"Adaptation (eye)","level":2,"score":0.44384193420410156},{"id":"https://openalex.org/C61328038","wikidata":"https://www.wikidata.org/wiki/Q3358061","display_name":"Speech processing","level":2,"score":0.2750924229621887},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.16912785172462463},{"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/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0},{"id":"https://openalex.org/C120665830","wikidata":"https://www.wikidata.org/wiki/Q14620","display_name":"Optics","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},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.21437/interspeech.2016-1010","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2016-1010","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2016","raw_type":"proceedings-article"},{"id":"pmh:oai:infoscience.epfl.ch:219850","is_oa":true,"landing_page_url":"http://infoscience.epfl.ch/record/219850","pdf_url":null,"source":{"id":"https://openalex.org/S4306400487","display_name":"Infoscience (Ecole Polytechnique F\u00e9d\u00e9rale de Lausanne)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I5124864","host_organization_name":"\u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne","host_organization_lineage":["https://openalex.org/I5124864"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"Text"}],"best_oa_location":{"id":"pmh:oai:infoscience.epfl.ch:219850","is_oa":true,"landing_page_url":"http://infoscience.epfl.ch/record/219850","pdf_url":null,"source":{"id":"https://openalex.org/S4306400487","display_name":"Infoscience (Ecole Polytechnique F\u00e9d\u00e9rale de Lausanne)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I5124864","host_organization_name":"\u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne","host_organization_lineage":["https://openalex.org/I5124864"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"Text"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":30,"referenced_works":["https://openalex.org/W308849112","https://openalex.org/W1510837697","https://openalex.org/W1526516369","https://openalex.org/W1534441687","https://openalex.org/W1553004968","https://openalex.org/W1590760721","https://openalex.org/W1978660892","https://openalex.org/W2025198378","https://openalex.org/W2027030707","https://openalex.org/W2029757240","https://openalex.org/W2033436836","https://openalex.org/W2045211645","https://openalex.org/W2051530898","https://openalex.org/W2054509039","https://openalex.org/W2065927558","https://openalex.org/W2098624096","https://openalex.org/W2103487404","https://openalex.org/W2106440210","https://openalex.org/W2123798005","https://openalex.org/W2125838338","https://openalex.org/W2131968858","https://openalex.org/W2141820854","https://openalex.org/W2143217142","https://openalex.org/W2152671307","https://openalex.org/W2158069733","https://openalex.org/W2167771929","https://openalex.org/W2210138121","https://openalex.org/W2401271578","https://openalex.org/W3150637114","https://openalex.org/W4253573210"],"related_works":["https://openalex.org/W2153098279","https://openalex.org/W2121652828","https://openalex.org/W2126322296","https://openalex.org/W2092406870","https://openalex.org/W2163537793","https://openalex.org/W4386794561","https://openalex.org/W2916997151","https://openalex.org/W2781555308","https://openalex.org/W3021690593","https://openalex.org/W2125343999"],"abstract_inverted_index":{"Development":[0],"of":[1,80,91,113,116,131,139],"state-of-the-art":[2],"automatic":[3,132],"speech":[4],"recognition":[5],"(ASR)":[6],"systems":[7],"requires":[8],"acoustic":[9,26,38,97],"resources":[10,18],"(i.e.,":[11,19],"transcribed":[12],"speech)":[13],"as":[14,16],"well":[15],"lexical":[17,28,52,99],"phonetic":[20],"lexicons).It":[21],"has":[22],"been":[23],"shown":[24],"that":[25,40,54,68,153],"and":[27,48,86,98,128],"resource":[29],"constraints":[30],"can":[31,72],"be":[32,73],"overcome":[33],"by":[34],"first":[35],"training":[36,50],"an":[37,70],"model":[39,53,100],"captures":[41,55],"acoustic-to-multilingual":[42],"phone":[43,57,119,126,142],"relationships":[44,58],"on":[45,59,122,137,145],"languageindependent":[46],"data;":[47],"then":[49],"a":[51,77,114,148],"grapheme-to-multilingual":[56],"the":[60,89,92,108,117,123,140,162,166],"target":[61],"language":[62],"data.In":[63],"this":[64],"paper,":[65],"we":[66,102],"show":[67,152],"such":[69],"approach":[71,164],"employed":[74],"to":[75,106,157],"discover":[76,107],"latent":[78,109],"space":[79,135],"subword":[81,133],"units":[82],"for":[83],"under-resourced":[84,150],"languages,":[85],"subsequently":[87],"improve":[88],"performance":[90,159],"ASR":[93],"system":[94],"through":[95],"both":[96,154],"adaptation.Specifically,":[101],"present":[103],"two":[104],"approaches":[105,155],"space:":[110],"(1)":[111],"inference":[112],"subset":[115],"multilingual":[118],"set":[120],"based":[121,136],"learned":[124],"graphemeto-multilingual":[125,141],"relationships,":[127],"(2)":[129],"derivation":[130],"unit":[134],"clustering":[138],"relationships.Experimental":[143],"studies":[144],"Scottish":[146],"Gaelic,":[147],"truly":[149],"language,":[151],"lead":[156],"significant":[158],"improvements,":[160],"with":[161],"latter":[163],"yielding":[165],"best":[167],"system.":[168]},"counts_by_year":[],"updated_date":"2026-08-13T07:04:57.449891","created_date":"2025-10-10T00:00:00"}