What counts as real? Dataset of mediated bonafide and spoofed recordings
Abstract
Deepfakes mixed dataset is a collection of Voice Quality modified MLAAD deepfake and real audio samples. The dataset is designed to facilitate research in the field of deepfake detection and voice quality analysis. It includes a variety of audio samples that have been modified using different voice quality such as creaky, breathy, end-creak etc., as well as unmodified real audio samples for comparison.
Fields of Study & Subjects
Speech Recognition and Synthesis
Digital Media Forensic Detection
Speech and Audio Processing
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 27 February 2026
- Language
- EN