%0 Conference Proceedings
%T Learning Robust Self-Attention Features for Speech Emotion Recognition with Label-Adaptive Mixup
%A Lei Kang
%A Lichao Zhang
%A Dazhi Jiang
%B IEEE International Conference on Acoustics, Speech and Signal Processing
%D 2023
%F Lei Kang2023
%O LAMP
%O exported from refbase (http://158.109.8.37/show.php?record=3984), last updated on Mon, 29 Jan 2024 10:18:00 +0100
%X Speech Emotion Recognition (SER) is to recognize human emotions in a natural verbal interaction scenario with machines, which is considered as a challenging problem due to the ambiguous human emotions. Despite the recent progress in SER, state-of-the-art models struggle to achieve a satisfactory performance. We propose a self-attention based method with combined use of label-adaptive mixup and center loss. By adapting label probabilities in mixup and fitting center loss to the mixup training scheme, our proposed method achieves a superior performance to the state-of-the-art methods.
%U https://ieeexplore.ieee.org/abstract/document/10095611
%U http://dx.doi.org/10.1109/ICASSP49357.2023.10095611