Improving Cross-Corpus Speech Emotion Recognition with Adversarial Discriminative Domain Generalization (ADDoG)

Understand

Automatic speech emotion recognition provides computers with critical context to enable user understanding.

• While methods trained and tested within the same dataset have been shown successful, they often fail when applied to unseen datasets.
• To address this, recent work has focused on adversarial methods to find more generalized representations of emotional speech.
• However, many of these methods have issues converging, and only involve datasets collected in laboratory conditions.