WaveScat: Wavelet Scattering Front-Ends with Self-Supervised Features for Speech Deepfake Detection
The WST-X family of feature extractors combines Wavelet Scattering Transform (WST) with self-supervised learning features for speech deepfake detection. This method creates interpretable, robust, and deformation-stable multi-scale features that are mathematically transparent. It moves beyond traditional filters by capturing complex acoustic details across different scales.