JunXueTech/

TFCL

This repository provides the paired speech data and metadata used by TFCL, a time-frequency consistency learning framework for robust speech deepfake detection under acoustic front-end distortions.

Paper: Time-Frequency Consistency Learning for Robust Speech Deepfake DetectionAuthors: Jun Xue, Zhuolin Yi, Yanzhen Ren, Yihuan Huang, Jiayu Xiong, Yi Chai, Guanxiang Feng, Jiajun Liu, and Tong ZhangCode: JunXue-tech/TFCL… See the full description on the dataset page: https://huggingface.co/datasets/JunXueTech/TFCL.