Adaptive Multi-Modal Transformer Decoders for Robust Speech and Scene Classification Using Audio– Visual Fusion. int. J. Eng. Res. Sci. Tech. 2026;22(1):516-521. doi:10.62643/