{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,6,19]],"date-time":"2025-06-19T05:06:34Z","timestamp":1750309594130,"version":"3.41.0"},"reference-count":46,"publisher":"Association for Computing Machinery (ACM)","issue":"5","license":[{"start":{"date-parts":[[2025,5,22]],"date-time":"2025-05-22T00:00:00Z","timestamp":1747872000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.acm.org\/publications\/policies\/copyright_policy#Background"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China NSFC","doi-asserted-by":"crossref","award":["62472325"],"award-info":[{"award-number":["62472325"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Multimedia Comput. Commun. Appl."],"published-print":{"date-parts":[[2025,5,31]]},"abstract":"<jats:p>Existing work finds it challenging for adversarial examples to transfer among different synthetic speech detectors because of cross-feature and cross-model. To enhance the transferability of adversarial examples, we propose a spectral saliency analysis method and gain insight into the underlying detection mechanisms of existing detectors for the first time. These insights offer an interpretable basis for why adversarial examples are challenging to transfer between synthetic speech detection models. Then we further propose a two-stage adversarial attack framework. Specifically, the first stage leverages insights into the model detection mechanism to design a random time-frequency masking module, the random offset module, and 1D convolution to generate transferable and robust adversarial examples. In the second stage, to mitigate the problem of obvious noise in the low-energy frames of the carrier in existing adversarial attacks, we perform secondary optimization on frames below the Signal-Noise-Rate threshold to enhance its auditory quality. Extensive experimental results demonstrate that the proposed method significantly enhances the transferability and robustness of adversarial examples, while simultaneously preserving the acoustic quality compared to typical approaches.<\/jats:p>","DOI":"10.1145\/3727341","type":"journal-article","created":{"date-parts":[[2025,4,1]],"date-time":"2025-04-01T16:09:41Z","timestamp":1743523781000},"page":"1-19","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["The Interpretable and Transferable Adversarial Attack against Synthetic Speech Detectors"],"prefix":"10.1145","volume":"21","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-0452-4031","authenticated-orcid":false,"given":"Jiacheng","family":"Deng","sequence":"first","affiliation":[{"name":"Key Laboratory of Aerospace Information Security and Trusted Computing, Ministry of Education, School of Cyber Science and Engineering, Wuhan University, Wuhan, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2510-9523","authenticated-orcid":false,"given":"Dengpan","family":"Ye","sequence":"additional","affiliation":[{"name":"Key Laboratory of Aerospace Information Security and Trusted Computing, Ministry of Education, School of Cyber Science and Engineering, Wuhan University, Wuhan, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-2198-7986","authenticated-orcid":false,"given":"Jizhi","family":"Li","sequence":"additional","affiliation":[{"name":"Key Laboratory of Aerospace Information Security and Trusted Computing, Ministry of Education, School of Cyber Science and Engineering, Wuhan University, Wuhan, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6796-650X","authenticated-orcid":false,"given":"Ziyi","family":"Liu","sequence":"additional","affiliation":[{"name":"Key Laboratory of Aerospace Information Security and Trusted Computing, Ministry of Education, School of Cyber Science and Engineering, Wuhan University, Wuhan, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9345-0312","authenticated-orcid":false,"given":"Long","family":"Tang","sequence":"additional","affiliation":[{"name":"Key Laboratory of Aerospace Information Security and Trusted Computing, Ministry of Education, School of Cyber Science and Engineering, Wuhan University, Wuhan, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3087-5580","authenticated-orcid":false,"given":"Yunming","family":"Zhang","sequence":"additional","affiliation":[{"name":"Key Laboratory of Aerospace Information Security and Trusted Computing, Ministry of Education, School of Cyber Science and Engineering, Wuhan University, Wuhan, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"320","published-online":{"date-parts":[[2025,5,22]]},"reference":[{"key":"e_1_3_1_2_2","doi-asserted-by":"publisher","DOI":"10.1109\/SP40001.2021.00004"},{"key":"e_1_3_1_3_2","doi-asserted-by":"publisher","DOI":"10.1109\/TIFS.2023.3345639"},{"key":"e_1_3_1_4_2","doi-asserted-by":"publisher","DOI":"10.1109\/TIFS.2023.3307908"},{"key":"e_1_3_1_5_2","first-page":"2196","volume-title":"Proceedings of the International Conference on Machine Learning. PMLR","author":"Croce Francesco","year":"2020","unstructured":"Francesco Croce and Matthias Hein. 2020. Minimally distorted adversarial examples with a fast adaptive boundary attack. In Proceedings of the International Conference on Machine Learning. PMLR, 2196\u20132205."},{"key":"e_1_3_1_6_2","unstructured":"Jiacheng Deng Shunyi Chen Li Dong Diqun Yan and Rangding Wang. 2022. Transferability of adversarial attacks on synthetic speech detection. arXiv:2205.07711. Retrieved from https:\/\/arxiv.org\/abs\/2205.07711"},{"key":"e_1_3_1_7_2","doi-asserted-by":"publisher","DOI":"10.1145\/3548606.3559357"},{"key":"e_1_3_1_8_2","doi-asserted-by":"publisher","DOI":"10.1109\/LSP.2022.3183538"},{"key":"e_1_3_1_9_2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00957"},{"key":"e_1_3_1_10_2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00444"},{"key":"e_1_3_1_11_2","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2014-443"},{"key":"e_1_3_1_12_2","unstructured":"Ian J. Goodfellow Jonathon Shlens and Christian Szegedy. 2014. Explaining and harnessing adversarial examples. arXiv:1412.6572. Retrieved from https:\/\/arxiv.org\/abs\/1412.6572"},{"key":"e_1_3_1_13_2","doi-asserted-by":"publisher","DOI":"10.1145\/3581783.3613841"},{"key":"e_1_3_1_14_2","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2022.3231480"},{"key":"e_1_3_1_15_2","doi-asserted-by":"publisher","DOI":"10.1109\/LSP.2021.3089437"},{"key":"e_1_3_1_16_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9747766"},{"key":"e_1_3_1_17_2","doi-asserted-by":"crossref","unstructured":"Piotr Kawa Marcin Plata and Piotr Syga. 2022. Defense against adversarial attacks on audio DeepFake detection. arXiv:2212.14597. Retrieved from https:\/\/arxiv.org\/abs\/2212.14597","DOI":"10.21437\/Interspeech.2023-409"},{"key":"e_1_3_1_18_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2022.109286"},{"key":"e_1_3_1_19_2","doi-asserted-by":"publisher","DOI":"10.1109\/IJCNN48605.2020.9207309"},{"key":"e_1_3_1_20_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2018.8462693"},{"key":"e_1_3_1_21_2","doi-asserted-by":"publisher","DOI":"10.1109\/PACRIM.1993.407206"},{"key":"e_1_3_1_22_2","doi-asserted-by":"publisher","DOI":"10.1201\/9781351251389-8"},{"key":"e_1_3_1_23_2","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2023.3285283"},{"key":"e_1_3_1_24_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.specom.2017.01.008"},{"key":"e_1_3_1_25_2","doi-asserted-by":"crossref","unstructured":"Michele Panariello Wanying Ge Hemlata Tak Massimiliano Todisco and Nicholas Evans. 2023. Malafide: A novel adversarial convolutive noise attack against deepfake and spoofing detection systems. arXiv:2306.07655. Retrieved from https:\/\/arxiv.org\/abs\/2306.07655","DOI":"10.21437\/Interspeech.2023-703"},{"key":"e_1_3_1_26_2","doi-asserted-by":"publisher","DOI":"10.1109\/JSTSP.2017.2684705"},{"key":"e_1_3_1_27_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP49357.2023.10096873"},{"key":"e_1_3_1_28_2","unstructured":"Yi Ren Chenxu Hu Xu Tan Tao Qin Sheng Zhao Zhou Zhao and Tie-Yan Liu. 2020. FastSpeech 2: Fast and high-quality end-to-end text to speech. arXiv:2006.04558. Retrieved from https:\/\/arxiv.org\/abs\/2006.04558"},{"key":"e_1_3_1_29_2","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2015-472"},{"key":"e_1_3_1_30_2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPRW59228.2023.00097"},{"key":"e_1_3_1_31_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP39728.2021.9414234"},{"key":"e_1_3_1_32_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.compeleceng.2022.108446"},{"key":"e_1_3_1_33_2","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2016-1066"},{"key":"e_1_3_1_34_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9746484"},{"key":"e_1_3_1_35_2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02297"},{"key":"e_1_3_1_36_2","doi-asserted-by":"publisher","DOI":"10.1109\/TIFS.2022.3222963"},{"key":"e_1_3_1_37_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.csl.2020.101114"},{"key":"e_1_3_1_38_2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00284"},{"key":"e_1_3_1_39_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9053747"},{"key":"e_1_3_1_40_2","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2023.3267610"},{"key":"e_1_3_1_41_2","unstructured":"Jiadi Yao Hong Luo and Xiao-Lei Zhang. 2023. Interpretable spectrum transformation attacks to speaker recognition. arXiv:2302.10686. Retrieved from https:\/\/arxiv.org\/abs\/2302.10686"},{"key":"e_1_3_1_42_2","doi-asserted-by":"publisher","DOI":"10.1109\/LSP.2023.3293429"},{"key":"e_1_3_1_43_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9746939"},{"key":"e_1_3_1_44_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP48485.2024.10448184"},{"key":"e_1_3_1_45_2","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM41043.2020.9155483"},{"key":"e_1_3_1_46_2","doi-asserted-by":"publisher","DOI":"10.1145\/3581783.3612475"},{"key":"e_1_3_1_47_2","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-2834"}],"container-title":["ACM Transactions on Multimedia Computing, Communications, and Applications"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3727341","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3727341","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,19]],"date-time":"2025-06-19T01:56:56Z","timestamp":1750298216000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3727341"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,5,22]]},"references-count":46,"journal-issue":{"issue":"5","published-print":{"date-parts":[[2025,5,31]]}},"alternative-id":["10.1145\/3727341"],"URL":"https:\/\/doi.org\/10.1145\/3727341","relation":{},"ISSN":["1551-6857","1551-6865"],"issn-type":[{"type":"print","value":"1551-6857"},{"type":"electronic","value":"1551-6865"}],"subject":[],"published":{"date-parts":[[2025,5,22]]},"assertion":[{"value":"2024-05-28","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-03-21","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-05-22","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}