2024:

  • Müller, N. M., Kawa, P., Hu, S., Williams, J., Sperl, P. & Böttinger, K. A New Approach to Voice Authenticity
    • Accepted to INTERSPEECH 2024
    • PDF
    • Topics: anti-spoofing, datasets, speech processing, audio processing
  • Müller, N. M., Kawa, P., Choong, W. H., Casanova, E., Gölge, E., Müller, T., Syga, P., Sperl, P. & Böttinger, K. MLAAD: The Multi-Language Audio Anti-Spoofing Dataset
    • Proceedings of IJCNN 2024
    • PDF
    • Topics: anti-spoofing, datasets, multi-language, speech processing, audio processing

2023:

  • Kawa, P., Plata, M., Czuba M., Szymański P., & Syga, P. Improved DeepFake Detection Using Whisper Features
    • Proceedings of INTERSPEECH 2023
    • PDF
    • Implementation
    • Topics: DeepFake detection, Whisper ASR, speech processing, audio processing
  • Kawa, P., Plata, M., & Syga, P. Defense against Adversarial Attacks on Audio DeepFake Detection
    • Proceedings of INTERSPEECH 2023
    • Demo
    • PDF
    • Implementation
    • Topics: DeepFake detection, speech processing, audio processing, adversarial attacks

2022:

  • Kawa, P., Plata, M., & Syga, P. SpecRNet: Towards Faster and More Accessible Audio DeepFake Detection
    • Proceedings of TrustCom 2022
    • PDF
    • Implementation
    • Topics: DeepFake detection, speech processing, audio processing
  • Kawa, P., Plata, M., & Syga, P. Attack Agnostic Dataset: Towards Generalization and Stabilization of Audio DeepFake
    • Proceedings of INTERSPEECH 2022
    • PDF
    • Implementation
    • Topics: neural networks generalization, DeepFake detection, speech processing, audio processing

2021:

  • Kawa, P., & Syga, P. Verify It Yourself: A Note on Activation Functions’ Influence on Fast DeepFake Detection
    • Proceedings of SECRYPT 2021
    • PDF
    • Topics: DeepFake detection, video processing, activation functions

2020:

  • Kawa, P., & Syga, P. A note on deepfake detection with low-resources
    • arXiv preprint
    • PDF
    • Tags: DeepFake detection, video processing, activation functions