RMER-DT: Robust multimodal emotion recognition in conversational contexts based on diffusion and transformers | Publicación