Matches in SemOpenAlex for { <https://semopenalex.org/work/W4328007740> ?p ?o ?g. }
- W4328007740 endingPage "187" @default.
- W4328007740 startingPage "179" @default.
- W4328007740 abstract "In recent years, significant progress has been made in deep model-based automatic speech recognition (ASR), leading to its widespread deployment in the real world. At the same time, adversarial attacks against deep ASR systems are highly successful. Various methods have been proposed to defend ASR systems from these attacks. However, existing classification based methods focus on the design of deep learning models while lacking exploration of domain specific features. This work leverages filter bank-based features to better capture the characteristics of attacks for improved detection. Furthermore, the paper analyses the potentials of using speech and non-speech parts separately in detecting adversarial attacks. In the end, considering adverse environments where ASR systems may be deployed, we study the impact of acoustic noise of various types and signal-to-noise ratios. Extensive experiments show that the inverse filter bank features generally perform better in both clean and noisy environments, the detection is effective using either speech or non-speech part, and the acoustic noise can largely degrade the detection performance." @default.
- W4328007740 created "2023-03-22" @default.
- W4328007740 creator A5049782518 @default.
- W4328007740 creator A5090108098 @default.
- W4328007740 date "2023-01-01" @default.
- W4328007740 modified "2023-09-27" @default.
- W4328007740 title "Leveraging Domain Features for Detecting Adversarial Attacks Against Deep Speech Recognition in Noise" @default.
- W4328007740 cites W2106822551 @default.
- W4328007740 cites W2148154194 @default.
- W4328007740 cites W2407023693 @default.
- W4328007740 cites W2586068394 @default.
- W4328007740 cites W2587891104 @default.
- W4328007740 cites W2772315375 @default.
- W4328007740 cites W2906211136 @default.
- W4328007740 cites W2954930777 @default.
- W4328007740 cites W2962747881 @default.
- W4328007740 cites W2963058500 @default.
- W4328007740 cites W2963161670 @default.
- W4328007740 cites W2963827914 @default.
- W4328007740 cites W2964301649 @default.
- W4328007740 cites W2969249185 @default.
- W4328007740 cites W2972678295 @default.
- W4328007740 cites W2973252307 @default.
- W4328007740 cites W3015241525 @default.
- W4328007740 cites W4220724967 @default.
- W4328007740 cites W4220922534 @default.
- W4328007740 doi "https://doi.org/10.1109/ojsp.2023.3256321" @default.
- W4328007740 hasPublicationYear "2023" @default.
- W4328007740 type Work @default.
- W4328007740 citedByCount "0" @default.
- W4328007740 crossrefType "journal-article" @default.
- W4328007740 hasAuthorship W4328007740A5049782518 @default.
- W4328007740 hasAuthorship W4328007740A5090108098 @default.
- W4328007740 hasBestOaLocation W43280077401 @default.
- W4328007740 hasConcept C100515483 @default.
- W4328007740 hasConcept C100675267 @default.
- W4328007740 hasConcept C105339364 @default.
- W4328007740 hasConcept C106131492 @default.
- W4328007740 hasConcept C108583219 @default.
- W4328007740 hasConcept C111919701 @default.
- W4328007740 hasConcept C115961682 @default.
- W4328007740 hasConcept C119857082 @default.
- W4328007740 hasConcept C120665830 @default.
- W4328007740 hasConcept C121332964 @default.
- W4328007740 hasConcept C134306372 @default.
- W4328007740 hasConcept C154945302 @default.
- W4328007740 hasConcept C163294075 @default.
- W4328007740 hasConcept C192209626 @default.
- W4328007740 hasConcept C204201278 @default.
- W4328007740 hasConcept C2776182073 @default.
- W4328007740 hasConcept C28490314 @default.
- W4328007740 hasConcept C31972630 @default.
- W4328007740 hasConcept C33923547 @default.
- W4328007740 hasConcept C36503486 @default.
- W4328007740 hasConcept C37736160 @default.
- W4328007740 hasConcept C41008148 @default.
- W4328007740 hasConcept C61328038 @default.
- W4328007740 hasConcept C76155785 @default.
- W4328007740 hasConcept C99498987 @default.
- W4328007740 hasConceptScore W4328007740C100515483 @default.
- W4328007740 hasConceptScore W4328007740C100675267 @default.
- W4328007740 hasConceptScore W4328007740C105339364 @default.
- W4328007740 hasConceptScore W4328007740C106131492 @default.
- W4328007740 hasConceptScore W4328007740C108583219 @default.
- W4328007740 hasConceptScore W4328007740C111919701 @default.
- W4328007740 hasConceptScore W4328007740C115961682 @default.
- W4328007740 hasConceptScore W4328007740C119857082 @default.
- W4328007740 hasConceptScore W4328007740C120665830 @default.
- W4328007740 hasConceptScore W4328007740C121332964 @default.
- W4328007740 hasConceptScore W4328007740C134306372 @default.
- W4328007740 hasConceptScore W4328007740C154945302 @default.
- W4328007740 hasConceptScore W4328007740C163294075 @default.
- W4328007740 hasConceptScore W4328007740C192209626 @default.
- W4328007740 hasConceptScore W4328007740C204201278 @default.
- W4328007740 hasConceptScore W4328007740C2776182073 @default.
- W4328007740 hasConceptScore W4328007740C28490314 @default.
- W4328007740 hasConceptScore W4328007740C31972630 @default.
- W4328007740 hasConceptScore W4328007740C33923547 @default.
- W4328007740 hasConceptScore W4328007740C36503486 @default.
- W4328007740 hasConceptScore W4328007740C37736160 @default.
- W4328007740 hasConceptScore W4328007740C41008148 @default.
- W4328007740 hasConceptScore W4328007740C61328038 @default.
- W4328007740 hasConceptScore W4328007740C76155785 @default.
- W4328007740 hasConceptScore W4328007740C99498987 @default.
- W4328007740 hasLocation W43280077401 @default.
- W4328007740 hasLocation W43280077402 @default.
- W4328007740 hasOpenAccess W4328007740 @default.
- W4328007740 hasPrimaryLocation W43280077401 @default.
- W4328007740 hasRelatedWork W1581699318 @default.
- W4328007740 hasRelatedWork W1995638733 @default.
- W4328007740 hasRelatedWork W2139175112 @default.
- W4328007740 hasRelatedWork W2248536413 @default.
- W4328007740 hasRelatedWork W2307514504 @default.
- W4328007740 hasRelatedWork W2418631473 @default.
- W4328007740 hasRelatedWork W2485008119 @default.
- W4328007740 hasRelatedWork W3171794711 @default.
- W4328007740 hasRelatedWork W3198667862 @default.
- W4328007740 hasRelatedWork W4386336391 @default.