{"id":"https://openalex.org/W4412944702","doi":"https://doi.org/10.18653/v1/2025.findings-acl.1190","title":"Lost in Transcription, Found in Distribution Shift: Demystifying Hallucination in Speech Foundation Models","display_name":"Lost in Transcription, Found in Distribution Shift: Demystifying Hallucination in Speech Foundation Models","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412944702","doi":"https://doi.org/10.18653/v1/2025.findings-acl.1190"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.findings-acl.1190","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.1190","pdf_url":"https://aclanthology.org/2025.findings-acl.1190.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: ACL 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-acl.1190.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5114337736","display_name":"Hanin Atwany","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hanin Atwany","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5022331909","display_name":"Abd\u00fcl Waheed","orcid":"https://orcid.org/0000-0001-5812-8822"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Abdul Waheed","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102775511","display_name":"Rita Singh","orcid":"https://orcid.org/0000-0003-3743-0162"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Rita Singh","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5008944385","display_name":"Monojit Choudhury","orcid":"https://orcid.org/0000-0001-7473-7839"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Monojit Choudhury","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5113017615","display_name":"Bhiksha Raj","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bhiksha Raj","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":28.3882,"has_fulltext":true,"cited_by_count":5,"citation_normalized_percentile":{"value":0.99448355,"is_in_top_1_percent":true,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":91,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"23181","last_page":"23203"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12946","display_name":"Fractal and DNA sequence analysis","score":0.6736999750137329,"subfield":{"id":"https://openalex.org/subfields/1312","display_name":"Molecular Biology"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},"topics":[{"id":"https://openalex.org/T12946","display_name":"Fractal and DNA sequence analysis","score":0.6736999750137329,"subfield":{"id":"https://openalex.org/subfields/1312","display_name":"Molecular Biology"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T14394","display_name":"Cognitive Science and Education Research","score":0.6427000164985657,"subfield":{"id":"https://openalex.org/subfields/2805","display_name":"Cognitive Neuroscience"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/foundation","display_name":"Foundation (evidence)","score":0.715978741645813},{"id":"https://openalex.org/keywords/transcription","display_name":"Transcription (linguistics)","score":0.6830849051475525},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5188289880752563},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.4592987596988678},{"id":"https://openalex.org/keywords/linguistics","display_name":"Linguistics","score":0.14664357900619507},{"id":"https://openalex.org/keywords/history","display_name":"History","score":0.13001129031181335},{"id":"https://openalex.org/keywords/philosophy","display_name":"Philosophy","score":0.07376378774642944},{"id":"https://openalex.org/keywords/archaeology","display_name":"Archaeology","score":0.06333479285240173}],"concepts":[{"id":"https://openalex.org/C2780966255","wikidata":"https://www.wikidata.org/wiki/Q5474306","display_name":"Foundation (evidence)","level":2,"score":0.715978741645813},{"id":"https://openalex.org/C179926584","wikidata":"https://www.wikidata.org/wiki/Q207714","display_name":"Transcription (linguistics)","level":2,"score":0.6830849051475525},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5188289880752563},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.4592987596988678},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.14664357900619507},{"id":"https://openalex.org/C95457728","wikidata":"https://www.wikidata.org/wiki/Q309","display_name":"History","level":0,"score":0.13001129031181335},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.07376378774642944},{"id":"https://openalex.org/C166957645","wikidata":"https://www.wikidata.org/wiki/Q23498","display_name":"Archaeology","level":1,"score":0.06333479285240173}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-acl.1190","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.1190","pdf_url":"https://aclanthology.org/2025.findings-acl.1190.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: ACL 2025","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-acl.1190","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.1190","pdf_url":"https://aclanthology.org/2025.findings-acl.1190.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: ACL 2025","raw_type":"proceedings-article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/10","display_name":"Reduced inequalities","score":0.4699999988079071}],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412944702.pdf","grobid_xml":"https://content.openalex.org/works/W4412944702.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2381393187","https://openalex.org/W2332779545","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W2028097510","https://openalex.org/W2505877856"],"abstract_inverted_index":{"Speech":[0],"foundation":[1],"models":[2,31,142],"trained":[3],"at":[4],"a":[5,43,130],"massive":[6],"scale,":[7],"both":[8,164],"in":[9,17,70,85,109,207],"terms":[10],"of":[11,21,138,192],"model":[12,119,122,204],"and":[13,34,51,92,121,166,174],"data":[14],"size,":[15,120],"result":[16],"robust":[18],"systems":[19],"capable":[20],"performing":[22],"multiple":[23],"speech":[24,28],"tasks,":[25],"including":[26],"automatic":[27],"recognition":[29],"(ASR).These":[30],"transcend":[32],"language":[33],"domain":[35],"barriers,":[36],"yet":[37],"effectively":[38],"measuring":[39],"their":[40],"performance":[41,62],"remains":[42],"challenge.Traditional":[44],"metrics":[45,197],"like":[46,169,198],"word":[47],"error":[48,53,127],"rate":[49,54,128],"(WER)":[50],"character":[52],"(CER)":[55],"are":[56],"commonly":[57],"used":[58],"to":[59,66,134,200],"evaluate":[60],"ASR":[61,110,141,203],"but":[63],"often":[64],"fail":[65],"reflect":[67],"transcription":[68],"quality":[69],"critical":[71],"contexts,":[72],"particularly":[73,206],"when":[74],"detecting":[75],"fabricated":[76],"outputs.This":[77],"phenomenon,":[78],"known":[79],"as":[80,89,116],"hallucination,":[81],"is":[82],"especially":[83],"concerning":[84],"high-stakes":[86,208],"domains":[87],"such":[88,115],"healthcare,":[90],"legal,":[91],"aviation,":[93],"where":[94],"errors":[95],"can":[96,150],"have":[97],"severe":[98],"consequences.In":[99],"our":[100],"work,":[101],"we":[102,132],"address":[103],"this":[104],"gap":[105],"by":[106],"investigating":[107],"hallucination":[108,126,153],"models.We":[111],"examine":[112],"how":[113],"factors":[114],"distribution":[117],"shifts,":[118],"architecture":[123],"influence":[124],"the":[125,190],"(HER),":[129],"metric":[131],"introduce":[133],"quantify":[135],"hallucinations.Our":[136],"analysis":[137],"over":[139],"20":[140],"reveals":[143],"three":[144],"key":[145],"insights:":[146],"(1)":[147],"High":[148],"WERs":[149,157],"mask":[151],"low":[152,156],"rates,":[154],"while":[155],"may":[158],"conceal":[159],"dangerous":[160],"hallucinations.(2)":[161],"Synthetic":[162],"noise,":[163,171],"adversarial":[165],"common":[167],"perturbations":[168],"white":[170],"pitch":[172],"shift,":[173],"time":[175],"stretching,":[176],"increase":[177],"HER.(3)":[178],"Distribution":[179],"shift":[180],"correlates":[181],"strongly":[182],"with":[183],"HER":[184,194],"(":[185],"=":[186],"0.91).Our":[187],"findings":[188],"highlight":[189],"importance":[191],"incorporating":[193],"alongside":[195],"traditional":[196],"WER":[199],"better":[201],"assess":[202],"performance,":[205],"domains.":[209]},"counts_by_year":[{"year":2026,"cited_by_count":4},{"year":2025,"cited_by_count":1}],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}