Matches in SemOpenAlex for { <https://semopenalex.org/work/W3091896612> ?p ?o ?g. }
Showing items 1 to 88 of
88
with 100 items per page.
- W3091896612 abstract "Despite their immense popularity, deep learning-based acoustic systems are inherently vulnerable to adversarial attacks, wherein maliciously crafted audios trigger target systems to misbehave. In this paper, we present SirenAttack, a new class of attacks to generate adversarial audios. Compared with existing attacks, SirenAttack highlights with a set of significant features: (i) versatile -- it is able to deceive a range of end-to-end acoustic systems under both white-box and black-box settings; (ii) effective -- it is able to generate adversarial audios that can be recognized as specific phrases by target acoustic systems; and (iii) stealthy -- it is able to generate adversarial audios indistinguishable from their benign counterparts to human perception. We empirically evaluate SirenAttack on a set of state-of-the-art deep learning-based acoustic systems (including speech command recognition, speaker recognition and sound event classification), with results showing the versatility, effectiveness, and stealthiness of SirenAttack. For instance, it achieves 99.45% attack success rate on the IEMOCAP dataset against the ResNet18 model, while the generated adversarial audios are also misinterpreted by multiple popular ASR platforms, including Google Cloud Speech, Microsoft Bing Voice, and IBM Speech-to-Text. We further evaluate three potential defense methods to mitigate such attacks, including adversarial training, audio downsampling, and moving average filtering, which leads to promising directions for further research." @default.
- W3091896612 created "2020-10-15" @default.
- W3091896612 creator A5033073212 @default.
- W3091896612 creator A5054310469 @default.
- W3091896612 creator A5058611515 @default.
- W3091896612 creator A5061219446 @default.
- W3091896612 creator A5068080767 @default.
- W3091896612 creator A5068885142 @default.
- W3091896612 date "2020-10-05" @default.
- W3091896612 modified "2023-10-18" @default.
- W3091896612 title "SirenAttack: Generating Adversarial Audio for End-to-End Acoustic Systems" @default.
- W3091896612 cites W1932198206 @default.
- W3091896612 cites W2019482303 @default.
- W3091896612 cites W2022217943 @default.
- W3091896612 cites W2038484192 @default.
- W3091896612 cites W2052666245 @default.
- W3091896612 cites W2103279261 @default.
- W3091896612 cites W2109364787 @default.
- W3091896612 cites W2113278353 @default.
- W3091896612 cites W2127141656 @default.
- W3091896612 cites W2194775991 @default.
- W3091896612 cites W2294068577 @default.
- W3091896612 cites W2526050071 @default.
- W3091896612 cites W2543927648 @default.
- W3091896612 cites W2549139847 @default.
- W3091896612 cites W2575585029 @default.
- W3091896612 cites W2593116425 @default.
- W3091896612 cites W2760938034 @default.
- W3091896612 cites W2791616807 @default.
- W3091896612 cites W2897865027 @default.
- W3091896612 cites W2902543210 @default.
- W3091896612 cites W2962826786 @default.
- W3091896612 cites W2962910554 @default.
- W3091896612 cites W2964137095 @default.
- W3091896612 cites W2964301649 @default.
- W3091896612 cites W4300824008 @default.
- W3091896612 doi "https://doi.org/10.1145/3320269.3384733" @default.
- W3091896612 hasPublicationYear "2020" @default.
- W3091896612 type Work @default.
- W3091896612 sameAs 3091896612 @default.
- W3091896612 citedByCount "48" @default.
- W3091896612 countsByYear W30918966122019 @default.
- W3091896612 countsByYear W30918966122020 @default.
- W3091896612 countsByYear W30918966122021 @default.
- W3091896612 countsByYear W30918966122022 @default.
- W3091896612 countsByYear W30918966122023 @default.
- W3091896612 crossrefType "proceedings-article" @default.
- W3091896612 hasAuthorship W3091896612A5033073212 @default.
- W3091896612 hasAuthorship W3091896612A5054310469 @default.
- W3091896612 hasAuthorship W3091896612A5058611515 @default.
- W3091896612 hasAuthorship W3091896612A5061219446 @default.
- W3091896612 hasAuthorship W3091896612A5068080767 @default.
- W3091896612 hasAuthorship W3091896612A5068885142 @default.
- W3091896612 hasConcept C108583219 @default.
- W3091896612 hasConcept C110384440 @default.
- W3091896612 hasConcept C115961682 @default.
- W3091896612 hasConcept C154945302 @default.
- W3091896612 hasConcept C177264268 @default.
- W3091896612 hasConcept C199360897 @default.
- W3091896612 hasConcept C28490314 @default.
- W3091896612 hasConcept C37736160 @default.
- W3091896612 hasConcept C41008148 @default.
- W3091896612 hasConceptScore W3091896612C108583219 @default.
- W3091896612 hasConceptScore W3091896612C110384440 @default.
- W3091896612 hasConceptScore W3091896612C115961682 @default.
- W3091896612 hasConceptScore W3091896612C154945302 @default.
- W3091896612 hasConceptScore W3091896612C177264268 @default.
- W3091896612 hasConceptScore W3091896612C199360897 @default.
- W3091896612 hasConceptScore W3091896612C28490314 @default.
- W3091896612 hasConceptScore W3091896612C37736160 @default.
- W3091896612 hasConceptScore W3091896612C41008148 @default.
- W3091896612 hasLocation W30918966121 @default.
- W3091896612 hasOpenAccess W3091896612 @default.
- W3091896612 hasPrimaryLocation W30918966121 @default.
- W3091896612 hasRelatedWork W2126887587 @default.
- W3091896612 hasRelatedWork W2731899572 @default.
- W3091896612 hasRelatedWork W2939353110 @default.
- W3091896612 hasRelatedWork W2962700793 @default.
- W3091896612 hasRelatedWork W3009238340 @default.
- W3091896612 hasRelatedWork W3091976719 @default.
- W3091896612 hasRelatedWork W3199202134 @default.
- W3091896612 hasRelatedWork W3215138031 @default.
- W3091896612 hasRelatedWork W4321369474 @default.
- W3091896612 hasRelatedWork W4360585206 @default.
- W3091896612 isParatext "false" @default.
- W3091896612 isRetracted "false" @default.
- W3091896612 magId "3091896612" @default.
- W3091896612 workType "article" @default.