{"id":"https://openalex.org/W3031277321","doi":"https://doi.org/10.21437/interspeech.2020-2629","title":"Improving Unsupervised Sparsespeech Acoustic Models with Categorical Reparameterization","display_name":"Improving Unsupervised Sparsespeech Acoustic Models with Categorical Reparameterization","publication_year":2020,"publication_date":"2020-10-25","ids":{"openalex":"https://openalex.org/W3031277321","doi":"https://doi.org/10.21437/interspeech.2020-2629","mag":"3031277321"},"language":"en","primary_location":{"id":"doi:10.21437/interspeech.2020-2629","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2020-2629","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2020","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2005.14578","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5078278300","display_name":"Benjamin Milde","orcid":null},"institutions":[{"id":"https://openalex.org/I31512782","display_name":"Technische Universit\u00e4t Darmstadt","ror":"https://ror.org/05n911h24","country_code":"DE","type":"education","lineage":["https://openalex.org/I31512782"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Benjamin Milde","raw_affiliation_strings":["Technical University of Darmstadt, Darmstadt, Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Technical University of Darmstadt, Darmstadt, Germany","institution_ids":["https://openalex.org/I31512782"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5021287757","display_name":"Chris Biemann","orcid":"https://orcid.org/0000-0002-8449-9624"},"institutions":[{"id":"https://openalex.org/I159176309","display_name":"Universit\u00e4t Hamburg","ror":"https://ror.org/00g30e956","country_code":"DE","type":"education","lineage":["https://openalex.org/I159176309"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Chris Biemann","raw_affiliation_strings":["Universit\u00e4t Hamburg, Hamburg, Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Universit\u00e4t Hamburg, Hamburg, Germany","institution_ids":["https://openalex.org/I159176309"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"2747","last_page":"2751"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":1.0,"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":1.0,"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/T10028","display_name":"Topic Modeling","score":0.9994000196456909,"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":0.9991000294685364,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/softmax-function","display_name":"Softmax function","score":0.7643601894378662},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7002894878387451},{"id":"https://openalex.org/keywords/categorical-variable","display_name":"Categorical variable","score":0.6552091240882874},{"id":"https://openalex.org/keywords/abx-test","display_name":"ABX test","score":0.5419806241989136},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5030307173728943},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.4885021448135376},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.4859234392642975},{"id":"https://openalex.org/keywords/connectionism","display_name":"Connectionism","score":0.4782808721065521},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4589672088623047},{"id":"https://openalex.org/keywords/backpropagation","display_name":"Backpropagation","score":0.4298696517944336},{"id":"https://openalex.org/keywords/acoustic-model","display_name":"Acoustic model","score":0.42011162638664246},{"id":"https://openalex.org/keywords/speech-processing","display_name":"Speech processing","score":0.2683511972427368},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.22703659534454346},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.16033312678337097},{"id":"https://openalex.org/keywords/statistics","display_name":"Statistics","score":0.1372905671596527}],"concepts":[{"id":"https://openalex.org/C188441871","wikidata":"https://www.wikidata.org/wiki/Q7554146","display_name":"Softmax function","level":3,"score":0.7643601894378662},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7002894878387451},{"id":"https://openalex.org/C5274069","wikidata":"https://www.wikidata.org/wiki/Q2285707","display_name":"Categorical variable","level":2,"score":0.6552091240882874},{"id":"https://openalex.org/C60962613","wikidata":"https://www.wikidata.org/wiki/Q287654","display_name":"ABX test","level":2,"score":0.5419806241989136},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5030307173728943},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.4885021448135376},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.4859234392642975},{"id":"https://openalex.org/C8521452","wikidata":"https://www.wikidata.org/wiki/Q203790","display_name":"Connectionism","level":3,"score":0.4782808721065521},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4589672088623047},{"id":"https://openalex.org/C155032097","wikidata":"https://www.wikidata.org/wiki/Q798503","display_name":"Backpropagation","level":3,"score":0.4298696517944336},{"id":"https://openalex.org/C155635449","wikidata":"https://www.wikidata.org/wiki/Q4674699","display_name":"Acoustic model","level":3,"score":0.42011162638664246},{"id":"https://openalex.org/C61328038","wikidata":"https://www.wikidata.org/wiki/Q3358061","display_name":"Speech processing","level":2,"score":0.2683511972427368},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.22703659534454346},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.16033312678337097},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.1372905671596527},{"id":"https://openalex.org/C13280743","wikidata":"https://www.wikidata.org/wiki/Q131089","display_name":"Geodesy","level":1,"score":0.0},{"id":"https://openalex.org/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.0}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.21437/interspeech.2020-2629","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2020-2629","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2020","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2005.14578","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2005.14578","pdf_url":"https://arxiv.org/pdf/2005.14578","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"public-domain","license_id":"https://openalex.org/licenses/public-domain","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"mag:3031277321","is_oa":true,"landing_page_url":"https://arxiv.org/pdf/2005.14578.pdf","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"arXiv (Cornell University)","raw_type":null},{"id":"doi:10.48550/arxiv.2005.14578","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2005.14578","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2005.14578","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2005.14578","pdf_url":"https://arxiv.org/pdf/2005.14578","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"public-domain","license_id":"https://openalex.org/licenses/public-domain","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[{"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4","score":0.5299999713897705}],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3031277321.pdf","grobid_xml":"https://content.openalex.org/works/W3031277321.grobid-xml"},"referenced_works_count":16,"referenced_works":["https://openalex.org/W10548402","https://openalex.org/W1494198834","https://openalex.org/W1524333225","https://openalex.org/W2127141656","https://openalex.org/W2395899413","https://openalex.org/W2520160253","https://openalex.org/W2547875792","https://openalex.org/W2593779438","https://openalex.org/W2842511635","https://openalex.org/W2940544976","https://openalex.org/W2947591107","https://openalex.org/W2963341956","https://openalex.org/W2972982164","https://openalex.org/W2979476256","https://openalex.org/W2995181338","https://openalex.org/W3015356564"],"related_works":["https://openalex.org/W3096301022","https://openalex.org/W2189391786","https://openalex.org/W1982117629","https://openalex.org/W2108868130","https://openalex.org/W2384074977","https://openalex.org/W3001745422","https://openalex.org/W2170477497","https://openalex.org/W2152192266","https://openalex.org/W2641129314","https://openalex.org/W2112909784","https://openalex.org/W2120731259","https://openalex.org/W2125610823","https://openalex.org/W2939164678","https://openalex.org/W2107135764","https://openalex.org/W2312118531","https://openalex.org/W1481751294","https://openalex.org/W3163479226","https://openalex.org/W1492282630","https://openalex.org/W2101191963","https://openalex.org/W2743367112"],"abstract_inverted_index":{"The":[0,32,77,99],"Sparsespeech":[1,19,152],"model":[2,7,20,45,81,100,115,153,166],"is":[3,82,101],"an":[4],"unsupervised":[5],"acoustic":[6],"that":[8],"can":[9,39],"generate":[10],"discrete":[11,28,59],"pseudo-labels":[12],"for":[13,23,92],"untranscribed":[14],"speech.":[15,110,129],"We":[16,49,111,130],"extend":[17],"the":[18,44,51,62,71,87,113,117,146,150,165],"to":[21,54,69,137,167],"allow":[22],"sampling":[24],"over":[25],"a":[26,58,90,122,132],"random":[27],"variable,":[29],"yielding":[30],"pseudo-posteriorgrams.":[31],"degree":[33],"of":[34,107,127,135,156],"sparsity":[35],"in":[36,61],"this":[37,66],"posteriorgram":[38],"be":[40],"fully":[41],"controlled":[42],"after":[43],"has":[46],"been":[47],"trained.":[48],"use":[50],"Gumbel-Softmax":[52],"trick":[53],"approximately":[55],"sample":[56],"from":[57],"distribution":[60],"neural":[63],"network":[64,72],"and":[65,79,84,105,121,159],"allows":[67],"us":[68],"train":[70],"efficiently":[73],"with":[74,94,125,149],"standard":[75],"backpropagation.":[76],"new":[78],"improved":[80,114,151],"trained":[83,102],"evaluated":[85],"on":[86,103,139,145,154],"Libri-Light":[88],"corpus,":[89],"benchmark":[91],"ASR":[93],"limited":[95],"or":[96],"no":[97],"supervision.":[98],"600h":[104,155],"6000h":[106],"English":[108],"read":[109],"evaluate":[112],"using":[116],"ABX":[118,140],"error":[119,141],"measure":[120],"semi-supervised":[123],"setting":[124],"10h":[126],"transcribed":[128],"observe":[131],"relative":[133],"improvement":[134],"up":[136],"31.4%":[138],"rates":[142],"across":[143],"speakers":[144],"test":[147],"set":[148],"speech":[157],"data":[158],"further":[160],"improvements":[161],"when":[162],"we":[163],"scale":[164],"6000h.":[168]},"counts_by_year":[{"year":2021,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}