🗞️ Paper “MemoCMT Multimodal Emotion Recognition Using Cross-Modal Transformer-Based Feature Fusion” was accepted by the Scientific Reports.

• Mustaqeem Khan^*, Phuong-Nam Tran^*, Nhat Truong Pham^*, Abdulmotaleb El Saddik, Alice Othmani, “MemoCMT: multimodal emotion recognition using cross-modal transformer-based feature fusion”, Scientific Reports, vol. 15, 5473, 2025 (^* Equal contribution)

MemoCMT recognizes emotions from speech and text together. It extracts audio features with HuBERT and text features with BERT, then fuses them with a cross-modal transformer. On the IEMOCAP and ESD benchmarks it reaches an unweighted accuracy of 81.33% and 91.93%, respectively. The code is available on GitHub.