Picture for Muhammad Sajidur Rahman

Muhammad Sajidur Rahman

Attacks as Defenses: Designing Robust Audio CAPTCHAs Using Attacks on Automatic Speech Recognition Systems

Add code
Mar 10, 2022
Figure 1 for Attacks as Defenses: Designing Robust Audio CAPTCHAs Using Attacks on Automatic Speech Recognition Systems
Figure 2 for Attacks as Defenses: Designing Robust Audio CAPTCHAs Using Attacks on Automatic Speech Recognition Systems
Figure 3 for Attacks as Defenses: Designing Robust Audio CAPTCHAs Using Attacks on Automatic Speech Recognition Systems
Figure 4 for Attacks as Defenses: Designing Robust Audio CAPTCHAs Using Attacks on Automatic Speech Recognition Systems
Viaarxiv icon

Beyond $L_p$ clipping: Equalization-based Psychoacoustic Attacks against ASRs

Add code
Oct 25, 2021
Figure 1 for Beyond $L_p$ clipping: Equalization-based Psychoacoustic Attacks against ASRs
Figure 2 for Beyond $L_p$ clipping: Equalization-based Psychoacoustic Attacks against ASRs
Figure 3 for Beyond $L_p$ clipping: Equalization-based Psychoacoustic Attacks against ASRs
Figure 4 for Beyond $L_p$ clipping: Equalization-based Psychoacoustic Attacks against ASRs
Viaarxiv icon

Hear "No Evil", See "Kenansville": Efficient and Transferable Black-Box Attacks on Speech Recognition and Voice Identification Systems

Add code
Oct 11, 2019
Figure 1 for Hear "No Evil", See "Kenansville": Efficient and Transferable Black-Box Attacks on Speech Recognition and Voice Identification Systems
Figure 2 for Hear "No Evil", See "Kenansville": Efficient and Transferable Black-Box Attacks on Speech Recognition and Voice Identification Systems
Figure 3 for Hear "No Evil", See "Kenansville": Efficient and Transferable Black-Box Attacks on Speech Recognition and Voice Identification Systems
Figure 4 for Hear "No Evil", See "Kenansville": Efficient and Transferable Black-Box Attacks on Speech Recognition and Voice Identification Systems
Viaarxiv icon