Advances in Authentication and Deepfake Detection

The field of authentication and deepfake detection is rapidly evolving, with a focus on developing more robust and reliable methods for verifying the authenticity of voices, images, and videos. Recent research has explored the use of inertial sensing, voice morphing, and deep learning techniques to improve the accuracy of authentication systems. One of the key challenges in this area is the ability to detect and prevent deepfakes, which are becoming increasingly sophisticated and realistic. To address this challenge, researchers are developing new methods for detecting deepfakes, including the use of scaling laws, deepfake detection frameworks, and triaging audio forgeries. Notable papers in this area include Beyond the Voice: Inertial Sensing of Mouth Motion for High Security Speech Verification, which presents a novel approach to speech verification using inertial sensing, and Scaling Laws for Deepfake Detection, which analyzes the performance of deepfake detection models and proposes a data-centric approach to improving their accuracy. Additionally, Fit for Purpose? Deepfake Detection in the Real World highlights the limitations of existing deepfake detection models and the need for more realistic and effective evaluation protocols.

Sources

Beyond the Voice: Inertial Sensing of Mouth Motion for High Security Speech Verification

VoiceMorph: How AI Voice Morphing Reveals the Boundaries of Auditory Self-Recognition

Scaling Laws for Deepfake Detection

Fit for Purpose? Deepfake Detection in the Real World

On the Credibility of Deniable Communication in Court

Not All Deepfakes Are Created Equal: Triaging Audio Forgeries for Robust Deepfake Singer Identification

Signature Forgery Detection: Improving Cross-Dataset Generalization

Online Handwritten Signature Verification Based on Temporal-Spatial Graph Attention Transformer

R2-SVC: Towards Real-World Robust and Expressive Zero-shot Singing Voice Conversion

Built with on top of