This is the official repository for the paper "MixFake: Benchmarking and Enhancing Audio Deepfake Detection in Diverse Real-world Mixed Audio".
2026/06/03 News: Our Code has been released.
2026/05/25 News✨: Our paper has been accepted as a Spotlight paper at ICME 2026.
2026/05/17 News: MixFake dataset is available at : HuggingFace
Run the scripts in sh/ to train or evaluate a model, for example:
sh sh/xxx.sh