Investigating self-supervised representations for audio-visual deepfake detection

Open in new window