Abstract
Negative emotions could increase the risks of traffic accidents. However, the driver’s emotional identification is rarely considered in the current design of intelligent vehicle alarm user terminals (IVAUTs). To solve the problem, this paper tries to design an IVAUT system based on emotional identification technology. Firstly, the transformer network was combined with a convolutional neural network (CNN) into a voice emotional identification system for intelligent vehicle alarm, and an emotional labeling approach was provided. Next, a bimodal fusion model was developed based on decompose-CNNs, which includes an appearance module, an optical flow module, and a bimodal fusion module. The proposed emotional identification approach was proved effective through experiments.
Highlights
Negative emotions could increase the risks of traffic accidents [1,2,3,4,5,6,7,8]
Hyperparameters δ, μ, and c are the weight of class center update rate, weight of loss function, and weight of encoder output, respectively. is paper designs a contrastive experiment to explore the influence of these hyperparameters on the voice emotion recognition effect. e voice emotion recognition accuracy under different hyperparameter settings is reported in Figure 6, where features A and B are Mel cepstral coefficient (MCC) and Mel frequency cepstral coefficient (MFCC), respectively
Conclusions is paper mainly develops an intelligent vehicle alarm user terminals (IVAUTs) system based on emotion identification technology
Summary
Negative emotions could increase the risks of traffic accidents [1,2,3,4,5,6,7,8]. To assist with driving, it is necessary to effectively detect the driver’s negative emotions and enhance his/her thinking ability, perception ability, and judgement ability [9,10,11]. With the popularity of intelligent vehicle interaction equipment, emotional identification technology has been gradually introduced to assist with driving by monitoring fatigue driving and driver emotions [19, 20]. A deep convolutional neural network (DCNN) was designed to recognize driver emotions, and an on-demand audio mechanism was developed to automatically collect audio resources with an online crawler, aiming to eliminate the driving risks induced by the driver’s negative emotions. E available techniques of voice emotion identification cannot effectively recognize the individual difference in voice, while the current methods of facial emotion identification overlook the correlation between appearance modal and optical flow modal.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
