Sun, Sep 6, 2026
| فارسی
1- Department of Computer Engineering، SR.C.، Islamic Azad University، Tehran، Iran
2- 2Department of Medical Radiation، Amirkabir University of Technology، Tehran، Iran
2- 2Department of Medical Radiation، Amirkabir University of Technology، Tehran، Iran
Abstract: (4 Views)
Abstract: Human speech is capable of conveying multiple emotions simultaneously, the accurate interpretation of which is important in understanding the complexity of human communication. Traditional speech emotion recognition systems usually detect only the dominant emotion or rely on multimodal inputs, such as visual and auditory cues. This study aims to use the Brain Emotion Learning (BEL) model to recognize overlapping and mixed emotional speech. Method: Overlapping emotional speech was synthesized based on the standard Berlin Emotional Speech Standard Database (EMO-DB) using the rich mathematical concept of convolution. The Berlin database includes seven emotional states: happiness, sadness, fear, disgust, boredom, anger, and neutral. First, the audio files were read by MATLAB software. Subsequently, twenty-one overlapping emotional states were generated by convolving two different emotional utterances spoken by the same speaker with t identical linguistic content. In the next step, the feature vector for each emotional speech was extracted. These vectors were then utilized as input for the brain emotional learning model. Finally, the blended emotion recognition accuracy of the proposed model was evaluated with common machine learning models. Findings: The results indicated that although the recognition accuracy of mixed emotional states with the brain emotional learning model is 30%, the Macro-F1 score of 53% demonstrates that at least one of the emotions present in the overlapping emotion is correctly identified to an acceptable extent. Conclusion: The brain emotional learning model in recognizing overlapped emotional speech provides a realistic reflection of the model’s performance in real-world conditions.
Keywords: Overlapping Emotional Speech, Mixed Emotional Speech, Brain Emotional Learning, Speech Emotional Learning
Type of Study: Research |
Received: 2026/08/11 | Accepted: 2026/08/27
Received: 2026/08/11 | Accepted: 2026/08/27
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
