A Novel DBN Feature Fusion Model for Cross-Corpus Speech Emotion Recognition

Figure 1

Spectrogram feature extraction process in SER.