{"id":"https://openalex.org/W4372260156","doi":"https://doi.org/10.1109/icassp49357.2023.10096788","title":"Do Coarser Units Benefit Cluster Prediction-Based Speech Pre-Training?","display_name":"Do Coarser Units Benefit Cluster Prediction-Based Speech Pre-Training?","publication_year":2023,"publication_date":"2023-05-05","ids":{"openalex":"https://openalex.org/W4372260156","doi":"https://doi.org/10.1109/icassp49357.2023.10096788"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49357.2023.10096788","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49357.2023.10096788","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":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://cnrs.hal.science/hal-04208427/document","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5091764449","display_name":"Ali Elkahky","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Ali Elkahky","raw_affiliation_strings":["Meta Inc","Meta AI (United States)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta Inc","institution_ids":[]},{"raw_affiliation_string":"Meta AI (United States)","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5051950818","display_name":"Wei-Ning Hsu","orcid":"https://orcid.org/0000-0001-5546-5217"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wei-Ning Hsu","raw_affiliation_strings":["Meta Inc","Meta AI (United States)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta Inc","institution_ids":[]},{"raw_affiliation_string":"Meta AI (United States)","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5027043104","display_name":"Paden Tomasello","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Paden Tomasello","raw_affiliation_strings":["Meta Inc","Meta AI (United States)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta Inc","institution_ids":[]},{"raw_affiliation_string":"Meta AI (United States)","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5112911727","display_name":"Tu Anh Nguyen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tu-Anh Nguyen","raw_affiliation_strings":["Meta Inc","ALMAnaCH - Automatic Language Modelling and ANAlysis & Computational Humanities (France)","Meta AI (United States)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta Inc","institution_ids":[]},{"raw_affiliation_string":"ALMAnaCH - Automatic Language Modelling and ANAlysis & Computational Humanities (France)","institution_ids":[]},{"raw_affiliation_string":"Meta AI (United States)","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5071809848","display_name":"Robin Algayres","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Robin Algayres","raw_affiliation_strings":["Meta Inc","CoML - Apprentissage machine et d\u00e9veloppement cognitif (France)","Meta AI (United States)","ALMAnaCH - Automatic Language Modelling and ANAlysis & Computational Humanities (France)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta Inc","institution_ids":[]},{"raw_affiliation_string":"CoML - Apprentissage machine et d\u00e9veloppement cognitif (France)","institution_ids":[]},{"raw_affiliation_string":"Meta AI (United States)","institution_ids":[]},{"raw_affiliation_string":"ALMAnaCH - Automatic Language Modelling and ANAlysis & Computational Humanities (France)","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5005191803","display_name":"Yossi Adi","orcid":"https://orcid.org/0000-0003-2237-3898"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yossi Adi","raw_affiliation_strings":["Meta Inc","Meta AI (United States)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta Inc","institution_ids":[]},{"raw_affiliation_string":"Meta AI (United States)","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5020182193","display_name":"Jade Copet","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jade Copet","raw_affiliation_strings":["Meta Inc","Meta AI (United States)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta Inc","institution_ids":[]},{"raw_affiliation_string":"Meta AI (United States)","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5007620149","display_name":"Emmanuel Dupoux","orcid":"https://orcid.org/0000-0002-7814-2952"},"institutions":[{"id":"https://openalex.org/I90669466","display_name":"\u00c9cole des hautes \u00e9tudes en sciences sociales","ror":"https://ror.org/02d9dg697","country_code":"FR","type":"facility","lineage":["https://openalex.org/I90669466"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Emmanuel Dupoux","raw_affiliation_strings":["Meta Inc","CoML - Apprentissage machine et d\u00e9veloppement cognitif (France)","EHESS - \u00c9cole des hautes \u00e9tudes en sciences sociales (54, boulevard Raspail 75006 Paris - France)","Meta AI (United States)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta Inc","institution_ids":[]},{"raw_affiliation_string":"CoML - Apprentissage machine et d\u00e9veloppement cognitif (France)","institution_ids":[]},{"raw_affiliation_string":"EHESS - \u00c9cole des hautes \u00e9tudes en sciences sociales (54, boulevard Raspail 75006 Paris - France)","institution_ids":["https://openalex.org/I90669466"]},{"raw_affiliation_string":"Meta AI (United States)","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101170706","display_name":"Abdelrahman Mohamed","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Abdelrahman Mohamed","raw_affiliation_strings":["Meta Inc","Meta AI (United States)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta Inc","institution_ids":[]},{"raw_affiliation_string":"Meta AI (United States)","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":9,"corresponding_author_ids":["https://openalex.org/A5091764449"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.5044,"has_fulltext":true,"cited_by_count":3,"citation_normalized_percentile":{"value":0.69798067,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":95},"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":0.9977999925613403,"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/T10028","display_name":"Topic Modeling","score":0.996399998664856,"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":0.8033298254013062},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.7517017126083374},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.6417232155799866},{"id":"https://openalex.org/keywords/cluster-analysis","display_name":"Cluster analysis","score":0.5639153122901917},{"id":"https://openalex.org/keywords/smoothing","display_name":"Smoothing","score":0.5436418056488037},{"id":"https://openalex.org/keywords/encoding","display_name":"Encoding (memory)","score":0.45758146047592163},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.45491743087768555},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.4130955934524536},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.4003259837627411},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.3547019362449646},{"id":"https://openalex.org/keywords/computer-vision","display_name":"Computer vision","score":0.0768950879573822}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8033298254013062},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.7517017126083374},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.6417232155799866},{"id":"https://openalex.org/C73555534","wikidata":"https://www.wikidata.org/wiki/Q622825","display_name":"Cluster analysis","level":2,"score":0.5639153122901917},{"id":"https://openalex.org/C3770464","wikidata":"https://www.wikidata.org/wiki/Q775963","display_name":"Smoothing","level":2,"score":0.5436418056488037},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.45758146047592163},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.45491743087768555},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.4130955934524536},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4003259837627411},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3547019362449646},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.0768950879573822},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0},{"id":"https://openalex.org/C187736073","wikidata":"https://www.wikidata.org/wiki/Q2920921","display_name":"Management","level":1,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/icassp49357.2023.10096788","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49357.2023.10096788","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"},{"id":"pmh:oai:HAL:hal-04208427v1","is_oa":true,"landing_page_url":"https://cnrs.hal.science/hal-04208427","pdf_url":"https://cnrs.hal.science/hal-04208427/document","source":{"id":"https://openalex.org/S4306402512","display_name":"HAL (Le Centre pour la Communication Scientifique Directe)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1294671590","host_organization_name":"Centre National de la Recherche Scientifique","host_organization_lineage":["https://openalex.org/I1294671590"],"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":"https://2023.ieeeicassp.org/","raw_type":"Conference papers"}],"best_oa_location":{"id":"pmh:oai:HAL:hal-04208427v1","is_oa":true,"landing_page_url":"https://cnrs.hal.science/hal-04208427","pdf_url":"https://cnrs.hal.science/hal-04208427/document","source":{"id":"https://openalex.org/S4306402512","display_name":"HAL (Le Centre pour la Communication Scientifique Directe)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1294671590","host_organization_name":"Centre National de la Recherche Scientifique","host_organization_lineage":["https://openalex.org/I1294671590"],"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":"https://2023.ieeeicassp.org/","raw_type":"Conference papers"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education","score":0.7300000190734863}],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4372260156.pdf","grobid_xml":"https://content.openalex.org/works/W4372260156.grobid-xml"},"referenced_works_count":50,"referenced_works":["https://openalex.org/W130754613","https://openalex.org/W1494198834","https://openalex.org/W2121227244","https://openalex.org/W2138615112","https://openalex.org/W2752796333","https://openalex.org/W2842511635","https://openalex.org/W2896457183","https://openalex.org/W2962784628","https://openalex.org/W2963799213","https://openalex.org/W2963979492","https://openalex.org/W2965373594","https://openalex.org/W2973049979","https://openalex.org/W2995181338","https://openalex.org/W3005680577","https://openalex.org/W3034999214","https://openalex.org/W3035524453","https://openalex.org/W3036601975","https://openalex.org/W3140429000","https://openalex.org/W3197580070","https://openalex.org/W3203140070","https://openalex.org/W3209059054","https://openalex.org/W3209984917","https://openalex.org/W4221161761","https://openalex.org/W4224875474","https://openalex.org/W4226278833","https://openalex.org/W4226507725","https://openalex.org/W4281492411","https://openalex.org/W4287854499","https://openalex.org/W4297808394","https://openalex.org/W4303649106","https://openalex.org/W4307680525","https://openalex.org/W4313182775","https://openalex.org/W4385245566","https://openalex.org/W4394671563","https://openalex.org/W6629717138","https://openalex.org/W6678277124","https://openalex.org/W6680970901","https://openalex.org/W6739901393","https://openalex.org/W6755207826","https://openalex.org/W6766673545","https://openalex.org/W6769311223","https://openalex.org/W6774314701","https://openalex.org/W6780218876","https://openalex.org/W6790356757","https://openalex.org/W6802045652","https://openalex.org/W6810047917","https://openalex.org/W6810461289","https://openalex.org/W6810531757","https://openalex.org/W6810673746","https://openalex.org/W6845951457"],"related_works":["https://openalex.org/W1978572805","https://openalex.org/W2383807498","https://openalex.org/W1997992934","https://openalex.org/W1987225439","https://openalex.org/W4238188170","https://openalex.org/W1950940422","https://openalex.org/W4283822356","https://openalex.org/W2129146436","https://openalex.org/W2032507829","https://openalex.org/W2147282173"],"abstract_inverted_index":{"The":[0,84],"research":[1],"community":[2],"has":[3],"produced":[4],"many":[5],"successful":[6],"self-supervised":[7,24,123],"speech":[8,94,100],"representation":[9],"learning":[10,25],"methods":[11],"over":[12],"the":[13,42,46,50,66,98],"past":[14],"few":[15],"years.":[16],"Discrete":[17],"units":[18,56,88,109],"have":[19],"been":[20],"utilized":[21],"in":[22],"various":[23],"frameworks,":[26],"such":[27],"as":[28],"VQ-VAE":[29],"[1],":[30],"wav2vec":[31],"2.0":[32],"[2],":[33],"Hu-BERT":[34],"[3],":[35],"and":[36,48,60,74,81,96],"Wav2Seq":[37],"[4].":[38],"This":[39],"paper":[40],"studies":[41],"impact":[43],"of":[44,52,69,86],"altering":[45],"granularity":[47],"improving":[49],"quality":[51,85],"these":[53],"discrete":[54],"acoustic":[55],"for":[57,112],"pre-training":[58],"encoder-only":[59,116],"encoder-decoder":[61,113],"models.":[62],"We":[63],"systematically":[64],"study":[65],"current":[67],"proposals":[68],"using":[70,92],"Byte-Pair":[71],"Encoding":[72],"(BPE)":[73],"new":[75],"extensions":[76],"that":[77,107],"use":[78],"cluster":[79],"smoothing":[80],"Brown":[82],"clustering.":[83],"learned":[87],"is":[89],"studied":[90],"intrinsically":[91],"zero":[93],"metrics":[95],"on":[97],"down-stream":[99],"recognition":[101],"(ASR)":[102],"task.":[103],"Our":[104],"results":[105],"suggest":[106],"longer-range":[108],"are":[110],"helpful":[111],"pre-training;":[114],"however,":[115],"masked-prediction":[117],"models":[118],"cannot":[119],"yet":[120],"benefit":[121],"from":[122],"word-like":[124],"targets.":[125]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":1}],"updated_date":"2026-05-07T13:39:58.223016","created_date":"2025-10-10T00:00:00"}