Analyzing Fairness in Deepfake Detection With Massively Annotated Databases | IEEE Journals & Magazine | IEEE Xplore