Dorothea Kolossa
Institution: TU Berlin
Research Hub(s):
Hub B: Embedded Security
E-Mail: dorothea.kolossa@rub.de
Publications:
Robustifying automatic speech recognition by extracting slowly varying features DistriBlock: Identifying adversarial audio samples by leveraging characteristics of the output distribution Dompteur: Taming Audio Adversarial Examples Detecting Adversarial Examples for Speech Recognition via Uncertainty Quantification Adversarial Attacks Against Automatic Speech Recognition Systems via Psychoacoustic Hiding Imperio: Robust Over-the-Air Adversarial Examples for Automatic Speech Recognition Systems Leveraging Frequency Analysis for Deep Fake Image Recognition
