{"id":"https://openalex.org/W7166853124","doi":"https://doi.org/10.18653/v1/2026.findings-acl.672","title":"MelTrim: Coarse-to-Fine Data Pruning for Speech Classification","display_name":"MelTrim: Coarse-to-Fine Data Pruning for Speech Classification","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7166853124","doi":"https://doi.org/10.18653/v1/2026.findings-acl.672"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2026.findings-acl.672","doi":"https://doi.org/10.18653/v1/2026.findings-acl.672","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.672","pdf_url":"https://aclanthology.org/2026.findings-acl.672.pdf","source":{"id":"https://openalex.org/S4363605144","display_name":"Findings of the Association for Computational Linguistics: ACL (Findings of ACL)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://aclanthology.org/2026.findings-acl.672.pdf","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5139733590","display_name":"Shaobo Wang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shaobo Wang","raw_affiliation_strings":["EPIC Lab , Shanghai Jiao Tong University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"EPIC Lab , Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114097318","display_name":"Tianle Niu","orcid":"https://orcid.org/0009-0008-0403-2903","observed_orcids":["https://orcid.org/0009-0008-0403-2903"]},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Tianle Niu","raw_affiliation_strings":["EPIC Lab , Shanghai Jiao Tong University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"EPIC Lab , Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125193264","display_name":"Xuan Ouyang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xuan Ouyang","raw_affiliation_strings":["EPIC Lab , Shanghai Jiao Tong University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"EPIC Lab , Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139802839","display_name":"Xintong Li","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xintong Li","raw_affiliation_strings":["EPIC Lab , Shanghai Jiao Tong University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"EPIC Lab , Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129646443","display_name":"Zhengkun Ge","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I12912129","display_name":"Northeastern University","ror":"https://ror.org/04t5xt781","country_code":"US","type":"education","lineage":["https://openalex.org/I12912129"]},{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]},{"id":"https://openalex.org/I9224756","display_name":"Northeastern University","ror":"https://ror.org/03awzbc87","country_code":"CN","type":"education","lineage":["https://openalex.org/I9224756"]}],"countries":["CN","US"],"is_corresponding":false,"raw_author_name":"Zhengkun Ge","raw_affiliation_strings":["EPIC Lab , Shanghai Jiao Tong University","School of Computer Science and Engineering , Northeastern University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"EPIC Lab , Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]},{"raw_affiliation_string":"School of Computer Science and Engineering , Northeastern University","institution_ids":["https://openalex.org/I12912129","https://openalex.org/I9224756"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129729439","display_name":"Yue Min","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yue Min","raw_affiliation_strings":["EPIC Lab , Shanghai Jiao Tong University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"EPIC Lab , Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139821971","display_name":"Xiaoqian Liu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I12912129","display_name":"Northeastern University","ror":"https://ror.org/04t5xt781","country_code":"US","type":"education","lineage":["https://openalex.org/I12912129"]},{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]},{"id":"https://openalex.org/I9224756","display_name":"Northeastern University","ror":"https://ror.org/03awzbc87","country_code":"CN","type":"education","lineage":["https://openalex.org/I9224756"]}],"countries":["CN","US"],"is_corresponding":false,"raw_author_name":"Xiaoqian Liu","raw_affiliation_strings":["EPIC Lab , Shanghai Jiao Tong University","School of Computer Science and Engineering , Northeastern University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"EPIC Lab , Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]},{"raw_affiliation_string":"School of Computer Science and Engineering , Northeastern University","institution_ids":["https://openalex.org/I12912129","https://openalex.org/I9224756"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103183873","display_name":"Hankun Wang","orcid":"https://orcid.org/0000-0003-0338-1809","observed_orcids":["https://orcid.org/0000-0003-0338-1809"]},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hankun Wang","raw_affiliation_strings":["EPIC Lab , Shanghai Jiao Tong University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"EPIC Lab , Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5139734154","display_name":"Linfeng Zhang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Linfeng Zhang","raw_affiliation_strings":["EPIC Lab , Shanghai Jiao Tong University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"EPIC Lab , Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.46729422,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"13751","last_page":"13765"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12535","display_name":"Machine Learning and Data Classification","score":0.8554999828338623,"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/T12535","display_name":"Machine Learning and Data Classification","score":0.8554999828338623,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.08659999817609787,"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/T10860","display_name":"Speech and Audio Processing","score":0.014000000432133675,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/dataset-pruning","display_name":"dataset pruning","score":0.9869999885559082},{"id":"https://openalex.org/keywords/speech-classification","display_name":"speech classification","score":0.7559999823570251},{"id":"https://openalex.org/keywords/coreset-construction","display_name":"coreset construction","score":0.7749999761581421},{"id":"https://openalex.org/keywords/mel-frequency-cepstral-coefficients","display_name":"Mel-frequency cepstral coefficients","score":0.7689999938011169},{"id":"https://openalex.org/keywords/umap-dimensionality-reduction","display_name":"UMAP dimensionality reduction","score":0.9739999771118164},{"id":"https://openalex.org/keywords/speech-emotion-recognition","display_name":"speech emotion recognition","score":0.8389999866485596},{"id":"https://openalex.org/keywords/speaker-identification","display_name":"speaker identification","score":0.9769999980926514}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.597000002861023},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.573199987411499},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.45489999651908875},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.4016999900341034},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.3846000134944916},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3465999960899353},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.3167000114917755},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.3043000102043152},{"id":"https://openalex.org/C12267149","wikidata":"https://www.wikidata.org/wiki/Q282453","display_name":"Support vector machine","level":2,"score":0.2556999921798706},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.2538999915122986}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.findings-acl.672","doi":"https://doi.org/10.18653/v1/2026.findings-acl.672","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.672","pdf_url":"https://aclanthology.org/2026.findings-acl.672.pdf","source":{"id":"https://openalex.org/S4363605144","display_name":"Findings of the Association for Computational Linguistics: ACL (Findings of ACL)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.findings-acl.672","doi":"https://doi.org/10.18653/v1/2026.findings-acl.672","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.672","pdf_url":"https://aclanthology.org/2026.findings-acl.672.pdf","source":{"id":"https://openalex.org/S4363605144","display_name":"Findings of the Association for Computational Linguistics: ACL (Findings of ACL)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[],"sustainable_development_goals_aurora":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166853124.pdf","grobid_xml":"https://content.openalex.org/works/W7166853124.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W2097360316","https://openalex.org/W4402112540","https://openalex.org/W7138238529","https://openalex.org/W139267174","https://openalex.org/W2136631194","https://openalex.org/W7164373610","https://openalex.org/W1590223631","https://openalex.org/W2153563431","https://openalex.org/W7214978415","https://openalex.org/W2168485018"],"abstract_inverted_index":{"Dataset":[0],"Pruning":[1],"(DP)":[2],"aims":[3,118],"to":[4,12,39,71,119,147,161],"construct":[5],"a":[6,54,66,162,180],"coreset":[7],"that":[8],"achieves":[9,159],"performance":[10,145],"comparable":[11],"the":[13,23,40,44,126,133,153,170,173,189],"original,":[14],"full":[15],"dataset.However,":[16],"few":[17],"studies":[18],"have":[19],"explored":[20],"DP":[21,150],"in":[22,42,75,100,165,179,183],"context":[24],"of":[25,128],"Speech":[26,140,154],"Classification":[27,141],"(SC)":[28],"tasks.Unlike":[29],"image":[30],"or":[31],"text":[32],"classification,":[33],"SC":[34],"is":[35,132,193],"particularly":[36],"challenging":[37],"due":[38],"difficulty":[41],"capturing":[43],"acoustic,":[45],"semantic,":[46],"and":[47,97],"contextual":[48],"representations.In":[49],"this":[50,131],"study,":[51],"we":[52,106],"propose":[53],"novel":[55],"dataset":[56,135],"pruning":[57,110,136],"method":[58,158],"for":[59,111,139,152],"speech":[60],"datasets,":[61],"termed":[62],"MelTrim,":[63],"which":[64,93,117],"uses":[65],"two-step":[67],"coarse-to-fine":[68],"framework":[69],"designed":[70,138],"address":[72],"these":[73],"challenges.Specifically,":[74],"Step":[76,104],"1,":[77],"MelTrim":[78],"coarsely":[79],"filters":[80],"utterance-level":[81],"redundant":[82],"samples":[83],"using":[84,102],"DB-SCAN":[85],"clustering":[86],"on":[87,169,188],"Mel-Frequency":[88],"Cepstral":[89],"Coefficients":[90],"(MFCC)":[91],"features,":[92],"are":[94],"first":[95,134],"flattened":[96],"then":[98],"reduced":[99],"dimensionality":[101],"UMAP.In":[103],"2,":[105],"perform":[107],"frame-level":[108],"redundancy":[109],"each":[112,124],"utterance":[113],"via":[114],"utility":[115],"pruning,":[116],"eliminate":[120],"irrelevant":[121],"frames":[122],"within":[123],"utterance.To":[125],"best":[127],"our":[129,157],"knowledge,":[130],"approach":[137],"tasks,":[142,176],"demonstrating":[143],"outstanding":[144],"compared":[146],"classical":[148],"general":[149],"methods.Notably,":[151],"Emotion":[155],"Recognition,":[156],"up":[160],"49.5%":[163],"improvement":[164],"WA":[166],"(Weighted":[167],"Accuracy)":[168],"MEAD":[171],"dataset.For":[172],"Speaker":[174],"Identification":[175],"it":[177],"results":[178],"41.9%":[181],"reduction":[182],"EER":[184],"(Equal":[185],"Error":[186],"Rate)":[187],"VoxCeleb1":[190],"dataset.Our":[191],"implementation":[192],"publicly":[194],"available":[195],"at":[196],"https:":[197],"//github.com/turturturturtur/MelTrim.":[198]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-07-02T00:00:00"}