learning discriminative features from spectrograms using center loss for speech emotion recognition
This work addresses the challenge of ambiguous emotion extraction in speech for improving machine-human interaction, representing an incremental advancement in feature learning methods.
The paper tackled the problem of extracting effective features for speech emotion recognition by proposing a novel approach that combines softmax cross-entropy loss and center loss to learn discriminative features from variable-length spectrograms, resulting in improvements of over 3% in unweighted and weighted accuracy on Mel-spectrogram input and over 4% on Short Time Fourier Transform spectrogram input.
Identifying the emotional state from speech is essential for the natural interaction of the machine with the speaker. However, extracting effective features for emotion recognition is difficult, as emotions are ambiguous. We propose a novel approach to learn discriminative features from variable length spectrograms for emotion recognition by cooperating softmax cross-entropy loss and center loss together. The softmax cross-entropy loss enables features from different emotion categories separable, and center loss efficiently pulls the features belonging to the same emotion category to their center. By combining the two losses together, the discriminative power will be highly enhanced, which leads to network learning more effective features for emotion recognition. As demonstrated by the experimental results, after introducing center loss, both the unweighted accuracy and weighted accuracy are improved by over 3\% on Mel-spectrogram input, and more than 4\% on Short Time Fourier Transform spectrogram input.