CFAD
CFAD: A Chinese Dataset for Fake Audio Detection
Install / Use
/learn @ADDchallenge/CFADREADME
README
CFAD Baseline Systems
Baseline systems are grouped by task:
- Fake Audio Detection
- Fake Algorithm Recognition
Please find more details from the paper.
For Fake Audio Detection Task:
Reference code is provided for the following systems:
- Baseline-LFCC-GMM (Python) LFCC feature extraction with GMM classifier
- Baseline-LFCC-LCNN (PyTorch) LFCC feature extraction with LCNN classifier (DNN)
- Baseline-RawNet2 (PyTorch)End-to-End DNN classifier
For Fake Algorithm Recognition Task:
Reference code is provided for the following systems:
- Baseline-LFCC-x-Vector (PyTorch) LFCC feature extraction with TDNN classifier (DNN)
- Baseline-LFCC-LCNN (PyTorch) LFCC feature extraction with LCNN classifier (DNN)
Dataset Download
The CFAD dataset is publicly available and can be downloaded from Zenodo https://zenodo.org/record/6635521\#.Ysjq4nZBw2x
