PET: High-Frequency Temporal Self-Consistency Learning for Partially Deepfake Audio Localization.

Jiayi He, Jiangyan Yi, Jianhua Tao, Siding Zeng

Browse the full ICASSP paper archive.