Audio-Textual Emotion Recognition Based on Improved Neural Networks

Figure 7

The confusion matrixes of single mode models: (a) model S-CBLA; (b) model T-BL.

(a)

(b)