{"id":"https://openalex.org/W4416249613","doi":"https://doi.org/10.1109/waspaa66052.2025.11230938","title":"Two Views, One Truth: Spectral and Self-Supervised Features Fusion for Robust Speech Deepfake Detection","display_name":"Two Views, One Truth: Spectral and Self-Supervised Features Fusion for Robust Speech Deepfake Detection","publication_year":2025,"publication_date":"2025-10-12","ids":{"openalex":"https://openalex.org/W4416249613","doi":"https://doi.org/10.1109/waspaa66052.2025.11230938"},"language":null,"primary_location":{"id":"doi:10.1109/waspaa66052.2025.11230938","is_oa":false,"landing_page_url":"https://doi.org/10.1109/waspaa66052.2025.11230938","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5072129532","display_name":"Yassine El Kheir","orcid":null},"institutions":[{"id":"https://openalex.org/I4210107233","display_name":"Language Science (South Korea)","ror":"https://ror.org/01h9v1373","country_code":"KR","type":"company","lineage":["https://openalex.org/I4210107233"]}],"countries":["KR"],"is_corresponding":true,"raw_author_name":"Yassine El Kheir","raw_affiliation_strings":["DFKI,Speech and Language Technology,Germany"],"affiliations":[{"raw_affiliation_string":"DFKI,Speech and Language Technology,Germany","institution_ids":["https://openalex.org/I4210107233"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103205617","display_name":"Arnab Das","orcid":"https://orcid.org/0000-0001-9715-1435"},"institutions":[{"id":"https://openalex.org/I4210107233","display_name":"Language Science (South Korea)","ror":"https://ror.org/01h9v1373","country_code":"KR","type":"company","lineage":["https://openalex.org/I4210107233"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Arnab Das","raw_affiliation_strings":["DFKI,Speech and Language Technology,Germany"],"affiliations":[{"raw_affiliation_string":"DFKI,Speech and Language Technology,Germany","institution_ids":["https://openalex.org/I4210107233"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5054749693","display_name":"Eren Erdogan","orcid":"https://orcid.org/0000-0001-7468-0617"},"institutions":[{"id":"https://openalex.org/I4210107233","display_name":"Language Science (South Korea)","ror":"https://ror.org/01h9v1373","country_code":"KR","type":"company","lineage":["https://openalex.org/I4210107233"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Enes Erdem Erdogan","raw_affiliation_strings":["DFKI,Speech and Language Technology,Germany"],"affiliations":[{"raw_affiliation_string":"DFKI,Speech and Language Technology,Germany","institution_ids":["https://openalex.org/I4210107233"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120354459","display_name":"Fabian Ritter-Guttierez","orcid":null},"institutions":[{"id":"https://openalex.org/I172675005","display_name":"Nanyang Technological University","ror":"https://ror.org/02e7b5302","country_code":"SG","type":"education","lineage":["https://openalex.org/I172675005"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Fabian Ritter-Guttierez","raw_affiliation_strings":["Nanyang Technological University,Singapore"],"affiliations":[{"raw_affiliation_string":"Nanyang Technological University,Singapore","institution_ids":["https://openalex.org/I172675005"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5034570279","display_name":"Tim Polzehl","orcid":"https://orcid.org/0000-0001-9592-0296"},"institutions":[{"id":"https://openalex.org/I4210107233","display_name":"Language Science (South Korea)","ror":"https://ror.org/01h9v1373","country_code":"KR","type":"company","lineage":["https://openalex.org/I4210107233"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Tim Polzehl","raw_affiliation_strings":["DFKI,Speech and Language Technology,Germany"],"affiliations":[{"raw_affiliation_string":"DFKI,Speech and Language Technology,Germany","institution_ids":["https://openalex.org/I4210107233"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5075001284","display_name":"Sebastian M\u00f6ller","orcid":"https://orcid.org/0000-0003-3057-0760"},"institutions":[{"id":"https://openalex.org/I4210107233","display_name":"Language Science (South Korea)","ror":"https://ror.org/01h9v1373","country_code":"KR","type":"company","lineage":["https://openalex.org/I4210107233"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Sebastian M\u00f6ller","raw_affiliation_strings":["DFKI,Speech and Language Technology,Germany"],"affiliations":[{"raw_affiliation_string":"DFKI,Speech and Language Technology,Germany","institution_ids":["https://openalex.org/I4210107233"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5072129532"],"corresponding_institution_ids":["https://openalex.org/I4210107233"],"apc_list":null,"apc_paid":null,"fwci":2.2283,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.91441899,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":96,"max":98},"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":0.4300000071525574,"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":0.4300000071525574,"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":0.13899999856948853,"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/T12357","display_name":"Digital Media Forensic Detection","score":0.09179999679327011,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/overfitting","display_name":"Overfitting","score":0.8309000134468079},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.7570000290870667},{"id":"https://openalex.org/keywords/fusion","display_name":"Fusion","score":0.5453000068664551},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.48919999599456787},{"id":"https://openalex.org/keywords/reduction","display_name":"Reduction (mathematics)","score":0.43560001254081726},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.34040001034736633}],"concepts":[{"id":"https://openalex.org/C22019652","wikidata":"https://www.wikidata.org/wiki/Q331309","display_name":"Overfitting","level":3,"score":0.8309000134468079},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.7570000290870667},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7177000045776367},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6345000267028809},{"id":"https://openalex.org/C158525013","wikidata":"https://www.wikidata.org/wiki/Q2593739","display_name":"Fusion","level":2,"score":0.5453000068664551},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.48919999599456787},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.43560001254081726},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.40540000796318054},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.34040001034736633},{"id":"https://openalex.org/C197424946","wikidata":"https://www.wikidata.org/wiki/Q1165717","display_name":"Waveform","level":3,"score":0.33739998936653137},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.32839998602867126},{"id":"https://openalex.org/C40969351","wikidata":"https://www.wikidata.org/wiki/Q3516228","display_name":"Word error rate","level":2,"score":0.3003999888896942},{"id":"https://openalex.org/C33954974","wikidata":"https://www.wikidata.org/wiki/Q486494","display_name":"Sensor fusion","level":2,"score":0.2912999987602234},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.2766000032424927}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/waspaa66052.2025.11230938","is_oa":false,"landing_page_url":"https://doi.org/10.1109/waspaa66052.2025.11230938","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":35,"referenced_works":["https://openalex.org/W2085662862","https://openalex.org/W2590129515","https://openalex.org/W3026777299","https://openalex.org/W3197358873","https://openalex.org/W3198329097","https://openalex.org/W3199131409","https://openalex.org/W3199161700","https://openalex.org/W3199956586","https://openalex.org/W3200319900","https://openalex.org/W3201197228","https://openalex.org/W3201397251","https://openalex.org/W3201773091","https://openalex.org/W3211424380","https://openalex.org/W4225527248","https://openalex.org/W4225534571","https://openalex.org/W4297841654","https://openalex.org/W4297841787","https://openalex.org/W4385822347","https://openalex.org/W4389524500","https://openalex.org/W4392910532","https://openalex.org/W4392910535","https://openalex.org/W4392910562","https://openalex.org/W4399793646","https://openalex.org/W4402111996","https://openalex.org/W4402112384","https://openalex.org/W4402112412","https://openalex.org/W4402115985","https://openalex.org/W4403955716","https://openalex.org/W4408353629","https://openalex.org/W4408354307","https://openalex.org/W4408354631","https://openalex.org/W4408799803","https://openalex.org/W4409487087","https://openalex.org/W4411120545","https://openalex.org/W4415707876"],"related_works":[],"abstract_inverted_index":{"Recent":[0],"advances":[1],"in":[2,44,155],"synthetic":[3],"speech":[4],"have":[5],"made":[6],"audio":[7,176],"deepfakes":[8],"increasingly":[9],"realistic,":[10],"posing":[11],"significant":[12],"security":[13],"risks.":[14],"Existing":[15],"detection":[16],"methods":[17],"that":[18,59,87,163],"rely":[19],"on":[20,121],"a":[21,104,151],"single":[22],"modality,":[23],"either":[24],"raw":[25],"waveform":[26,167],"embeddings":[27],"or":[28],"spectral-based":[29],"features,":[30],"are":[31],"vulnerable":[32],"to":[33,39,47,108],"non-spoof":[34],"disturbances":[35],"and":[36,75,103,130,168],"often":[37],"overfit":[38],"known":[40],"forgery":[41],"algorithms,":[42],"resulting":[43],"poor":[45],"generalization":[46,132,149],"unseen":[48],"attacks.":[49],"To":[50],"address":[51],"these":[52,81],"shortcomings,":[53],"we":[54],"investigate":[55],"hybrid":[56],"fusion":[57,82,95,135],"frameworks":[58],"integrate":[60],"self-supervised":[61],"learning":[62],"(SSL)-based":[63],"representations":[64,174],"with":[65,113,142,150],"handcrafted":[66],"spectral":[67,115,169],"descriptors":[68],"(e.g.,":[69],"MFCC,":[70],"LFCC,":[71],"CQCC).":[72],"By":[73],"aligning":[74],"combining":[76],"complementary":[77],"information":[78],"across":[79],"modalities,":[80],"approaches":[83,89],"capture":[84],"subtle":[85],"artifacts":[86],"single-feature":[88],"typically":[90],"overlook.":[91],"We":[92,117],"explore":[93],"several":[94],"strategies,":[96],"including":[97],"simple":[98],"concatenation,":[99],"cross-attention,":[100,102],"mutual":[101],"learnable":[105],"gating":[106],"mechanism,":[107],"optimally":[109],"blend":[110],"SSL":[111],"features":[112],"fine-grained":[114],"cues.":[116],"evaluate":[118],"our":[119],"approach":[120],"four":[122],"challenging":[123],"public":[124],"benchmarks":[125],"(LA19,":[126],"DF21,":[127],"ITW,":[128],"ASV5)":[129],"report":[131],"performance.":[133],"All":[134],"variants":[136],"consistently":[137],"outperform":[138],"an":[139],"SSL-only":[140],"baseline,":[141],"the":[143,147],"cross-attention":[144],"strategy":[145],"achieving":[146],"best":[148],"38%":[152],"relative":[153],"reduction":[154],"equal":[156],"error":[157],"rate":[158],"(EER).":[159],"These":[160],"results":[161],"confirm":[162],"joint":[164],"modeling":[165],"of":[166],"views":[170],"produces":[171],"robust,":[172],"domain-agnostic":[173],"for":[175],"deepfake":[177],"detection.":[178]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-04-21T08:09:41.155169","created_date":"2025-11-14T00:00:00"}