A deep neural network model using convolutional neural networks (CNN) and acoustic feature extraction was developed to classify audio datasets into seven emotions for detecting stress and anxiety.
A deep neural network model using CNN and acoustic feature extraction can classify audio into seven emotions to detect stress and anxiety.
Emotional strain is referred to as stress. Both our mental health and the mental health of others around us may be affected. While anxiety is a normal response to stress that may be frightening, it can also cause panic attacks. Everyone needs to address these mental health problems. The method we use to identify stress and anxiety in a person using vocal/audio information is described in this study. We have created a deep neural network model for stress and anxiety detection. Here, Kaggle audio datasets with 7 different emotions―joy, fear, disgust, neutral, sorrow, surprise, and anger―are taken into consideration. To train and evaluate classification algorithms like CNN, these audio datasets are employed. The audio is then pre-processed using acoustic feature extraction, and CNN is used to classify it, providing accuracy based on those seven emotions. This allows us to foretell if the individual is stressed out or anxious.
Mohapatra et al. (Thu,) conducted a other in Stress and anxiety. Deep neural network model (CNN) using vocal/audio information was evaluated on Accuracy of classifying 7 different emotions to detect stress and anxiety. A deep neural network model using convolutional neural networks (CNN) and acoustic feature extraction was developed to classify audio datasets into seven emotions for detecting stress and anxiety.