All's well that FID's well? Result quality and metric scores in GAN models for lip-sychronization tasks

Open in new window