IoT-Enabled WBAN and Machine Learning for Emotion Recognition

All industries, including computer science and health care and customer service are in dire need of efficient recognition of human feelings. This work is a script for a novel approach to affect recognition which involves both vocal emotion analysis and face emotion detection. Voice sentiment analysis project used machine learning technique to identify the prospects of emotional content in voice data. The model also assists in the classification of emotions with reference to the speech emotional states based on added features of MFCC and chroma. On the other hand, face emotion recognition project by applying deep learning algorithms identifies emotions by face expressions. With the help of convolutional, neural networks (CNNs) determining the happiness, sadness, and anger it analyses the facial features correctly. Combining all of these results in a coherent system that allows for the recognition of feelings through voice and or facial signals. Consequently, as demonstrated in the experimental results provided hereinafter, the accuracy of the presented system in correctly identifying human emotions is significantly higher than when using single models on their own. Therefore this combination creates space for other uses such as healthcare marketing among other fields hence improving user experience while at the same time providing unforeseen analysis of human conduct during an interaction with intelligent devices.