Work

Research on Speech Emotion Recognition Based on Machine Learning Approaches

Year: 2023
Type: dissertation
Abstract: With the continuous development of deep learning technology, research on humanmachine emotional interaction has made significant progress.Speech is one of the critical means of information transmission, playing an indispensable role in human life by conveying not only semantic but also emotional information.In recent years, emotion recognition based on speech has received much attention due to the widespread application of deep learning technology and the rise of affective computing.Accurately extracting features using the attention mechanism to capture the correspondence between the two features and enhance the contextual information of the fused features.In the comparison experiments, the Dual-TBNet model achieved a recognition accuracy and F1 score of 95.7% and 95.8% on the CASIA dataset, 66.7% and 65.6% on the eNTERFACE05 dataset, 64.8% and 64.9% on the IEMOCAP dataset, 84.1% and 84.3% on the EMO-DB dataset and 83.3% and 82.1% on the SAVEE dataset.The Dual-TBNet model effectively fuses acoustic features of different lengths and dimensions with pre-trained features, enhancing the robustness of the features, and achieved the best performance.
Author: Liu, Zheng
Language: English
Open Access status: green
OpenAlex ID: W7144595382
Cites: 118
Cited by:
Related to: 10