Data augmentation and enhancement for multimodal speech emotion recognition

Humans’ fundamental need is interaction with each other such as using conversation or speech. Therefore, it crucial to analyze speech computer technology determine emotions. The emotion recognition (SER) method detects emotions in by examining various aspects. SER a supervised decide the class This research proposed multimodal model one of deep learning based enhancement techniques, which attention mechanism. Additionally, this addresses imbalanced dataset problem field generative adversarial networks (GAN) data augmentation technique. achieved an excellent evaluation performance 0.96 96% for GAN configuration. work showed that could enhance and create balanced dataset.