THE USE OF CONDITIONAL GENERATIVE ADVERSARIAL NETWORKS FOR THE SYNTHESIS OF FACE IMAGES FROM SPEECH. ijmrr 2024, 14 (7), 92-99.