Dynamic relationship between oral English pronunciation standard and mental health monitored by biosensor
Xuan Zhou, Wei Jia, Cuiping Shi · Molecular & cellular biomechanics · 2024
Oral English pronunciation is an important feature of language ability, especially among non-native speakers, as good pronunciation has a direct impact on communication efficacy and social integration. However, the difficulties connected with attaining a high standard of oral English pronunciation can lead to severe stress, anxiety, and other mental health disorders. The purpose of the research is to establish a dynamic correlation between oral English pronunciation standards and mental health, as monitored through biosensor data. The research aims to explore how variations in speech accuracy and fluency during English pronunciation tasks can reflect underlying psychological states, such as stress, anxiety, and overall emotional well-being. The study proposed a novel Improved Flower Pollination-tuned Resilient Deep Neural Network (IFP-RDNN) in this article, to predict the oral English pronunciation rating using biosensors. Electroencephalography (EEG)records patterns of cerebral waves using electrodes applied to the head to assess the electrical impulses in the cerebellum called EEG signals was acquired during the listening state and with the audio signals utilized in stimuli, as well as the spoken audio obtained from the subject. The data processing used a median filter to remove noise from the audio data. Fast Fourier transform (FFT) is used to extract the features from the preprocessed data. It is measured by biomedical data, can be predicted with the help of an optimization technique which draws inspiration called IFP helps to optimize the parameters effectively by mimicking natural pollination processes; RDNN is employed with the optimized parameters; it can predict oral English pronunciation ratings. Experimental results reveal that the spoken audio confirms the improvement in pronunciation throughout the trials. In a comparative analysis, the suggested method is assessed with various evaluation measures, such as F1-score (88.9%), recall (91.60%), precision (89.80%), and accuracy (90.3%). The result demonstrated the IFP-RDNN method to predict the oral English pronunciation rating using biosensors. The findings indicate a significant connection between the quality of oral English pronunciation and mental health, with deviations from standard pronunciation being associated with increased stress and emotional suffering.