基于无监督图对比学习的语音情感识别

张雪梅,孙颖,张雪英

Speech emotion recognition with unsupervised graph contrastive learning

Xuemei ZHANG,Ying SUN,Xueying ZHANG

表 1 不同增强策略与池化方式下的模型对比

Tab.1 Performance comparison of models under different augmentation strategies and pooling methods

模型 IEMOCAP EMO-DB UA/% WA/% UA/% WA/% 模型1 60.86 61.23 88.16 87.01 模型2 63.42 63.49 89.10 87.69 模型3 63.94 64.04 89.45 88.48 模型4 68.84 69.05 90.50 89.68 SERUGCL 69.96 70.24 91.04 90.29