This dataset contains 1,866 audio samples (933 real, 933 synthetic) for training deepfake audio detection models. It is specifically designed for binary classification tasks to distinguish between authentic human speech and AI-generated synthetic audio.
