Explainable Lightweight Compact Deep Models for Speech Emotion Recognition
SoundArtificial IntelligenceEmerging TechnologiesMachine Learning
Summary
The authors developed a simpler and easier-to-understand model for recognizing emotions in speech, which is useful in areas like healthcare and customer service. Their method uses a small convolutional neural network that looks at speech patterns in a way humans can interpret, highlighting important parts of the speech that relate to emotions. They also use a visualization technique to show which parts of the speech inspired the model's decisions. Tests on a standard emotion speech dataset showed their approach works well while being more efficient and transparent than many complex existing methods.
Authors
Nelly Elsayed
Abstract
Speech Emotion Recognition (SER) is an important component in a wide range of human-centered applications, including healthcare, customer service, and human-omputer interaction. In medical and decision-support settings, there is increasing interest in models that not only achieve accurate emotion recognition but also support transparent predictions and efficient deployment. However, many existing SER approaches rely on complex deep learning architectures that limit interpretability and increase computational cost. This paper presents an explainable and lightweight speech emotion recognition framework based on a compact convolutional neural network architecture. The proposed approach utilizes log-Mel spectrogram representations to capture spectro-temporal speech characteristics and employs attentive statistics pooling to emphasize emotionally salient temporal segments. To improve model transparency, gradient-based class activation mapping (Grad-CAM) is incorporated to visualize the time-frequency regions that influence the model's predictions. Experimental evaluation on the SAVEE emotional speech dataset demonstrates that the proposed framework achieves competitive recognition performance while maintaining a compact architecture with significantly fewer parameters than many existing SER models. The results indicate that efficient convolutional architectures combined with interpretable analysis can provide a practical balance between recognition accuracy, computational efficiency, and model transparency.