{"id":"https://openalex.org/W4416035885","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.202","title":"MetaMixSpeech: Meta Task Augmentation for Low-Resource Speech Recognition","display_name":"MetaMixSpeech: Meta Task Augmentation for Low-Resource Speech Recognition","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4416035885","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.202"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.202","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.202","pdf_url":"https://aclanthology.org/2025.findings-emnlp.202.pdf","source":null,"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: EMNLP 2025","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.findings-emnlp.202.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101560152","display_name":"Yaqi Chen","orcid":"https://orcid.org/0000-0001-6301-802X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yaqi Chen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5024433209","display_name":"Hao Zhang","orcid":"https://orcid.org/0000-0001-8852-6311"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hao Zhang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100635256","display_name":"Wenlin Zhang","orcid":"https://orcid.org/0000-0002-6842-9722"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wenlin Zhang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114222670","display_name":"XuKui Yang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"XuKui Yang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5001488518","display_name":"Dan Qu","orcid":"https://orcid.org/0000-0001-9917-7794"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dan Qu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5100331837","display_name":"Yunpeng Liu","orcid":"https://orcid.org/0000-0002-1020-0191"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yunpeng Liu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"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":0.30372869,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"3769","last_page":"3779"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.6960999965667725,"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":0.6960999965667725,"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":0.12470000237226486,"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/T11309","display_name":"Music and Audio Processing","score":0.027699999511241913,"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/task","display_name":"Task (project management)","score":0.5645999908447266},{"id":"https://openalex.org/keywords/speaker-recognition","display_name":"Speaker recognition","score":0.33799999952316284},{"id":"https://openalex.org/keywords/speech-technology","display_name":"Speech technology","score":0.3059000074863434},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.26750001311302185}],"concepts":[{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.598800003528595},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.5645999908447266},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5302000045776367},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.3384999930858612},{"id":"https://openalex.org/C133892786","wikidata":"https://www.wikidata.org/wiki/Q1145189","display_name":"Speaker recognition","level":2,"score":0.33799999952316284},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3100999891757965},{"id":"https://openalex.org/C504749915","wikidata":"https://www.wikidata.org/wiki/Q9010971","display_name":"Speech technology","level":3,"score":0.3059000074863434},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.29109999537467957},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.26750001311302185},{"id":"https://openalex.org/C46312422","wikidata":"https://www.wikidata.org/wiki/Q11024","display_name":"Communication","level":1,"score":0.257999986410141},{"id":"https://openalex.org/C204201278","wikidata":"https://www.wikidata.org/wiki/Q1332614","display_name":"Voice activity detection","level":3,"score":0.2556999921798706}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-emnlp.202","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.202","pdf_url":"https://aclanthology.org/2025.findings-emnlp.202.pdf","source":null,"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: EMNLP 2025","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.202","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.202","pdf_url":"https://aclanthology.org/2025.findings-emnlp.202.pdf","source":null,"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: EMNLP 2025","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G2727895085","display_name":null,"funder_award_id":"232300421240","funder_id":"https://openalex.org/F4320323845","funder_display_name":"Natural Science Foundation of Henan Province"},{"id":"https://openalex.org/G7804056170","display_name":null,"funder_award_id":"62171470","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320323845","display_name":"Natural Science Foundation of Henan Province","ror":null}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416035885.pdf","grobid_xml":"https://content.openalex.org/works/W4416035885.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Meta-learning":[0],"has":[1],"proven":[2],"to":[3,41,43,126],"be":[4],"a":[5,116],"powerful":[6],"paradigm":[7],"for":[8],"effectively":[9,129],"improving":[10],"the":[11,51,89,96,106,131],"performance":[12],"of":[13],"low-resource":[14],"speech":[15],"recognition":[16],"by":[17,83],"learning":[18,26],"generalizable":[19],"knowledge":[20],"across":[21,138],"multiple":[22],"tasks.However,":[23],"multilingual":[24],"meta":[25],"also":[27],"faces":[28],"challenges":[29],"such":[30],"as":[31],"task":[32,53,67,81],"overfitting":[33,132],"and":[34,60,77,91,94,108,134,141],"learner":[35],"overfitting,":[36],"thereby":[37],"reducing":[38],"its":[39],"ability":[40],"generalize":[42],"new":[44],"tasks.To":[45],"address":[46],"these":[47],"issues,":[48],"we":[49,63],"augment":[50],"metatraining":[52],"with":[54],"\"more":[55],"data\"":[56],"during":[57],"both":[58,74],"training":[59],"evaluation":[61],"phases.Concretely,":[62],"propose":[64],"an":[65],"interpolation-based":[66],"augmentation":[68,76],"method":[69],"called":[70],"MetaMixSpeech,":[71],"which":[72],"includes":[73],"support":[75,90],"query":[78,92],"augmentation.MetaMixSpeech":[79],"enhances":[80],"diversity":[82],"linearly":[84],"combining":[85],"perturbed":[86],"features":[87],"from":[88],"sets":[93],"performing":[95],"same":[97],"linear":[98],"interpolation":[99],"on":[100,105],"their":[101],"corresponding":[102],"losses.Experimental":[103],"results":[104],"FLEURS":[107],"Common":[109],"Voice":[110],"datasets":[111,140],"demonstrate":[112],"that":[113],"MetaMixSpeech":[114],"achieves":[115],"6.35":[117],"%":[118],"improvement":[119],"in":[120],"Word":[121],"Error":[122],"Rate":[123],"(WER)":[124],"compared":[125],"meta-learning":[127],"approaches,":[128],"mitigating":[130],"problem":[133],"showcasing":[135],"superior":[136],"generalization":[137],"diverse":[139],"language":[142],"families.":[143]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-11-08T00:00:00"}