<![CDATA[The pandemic has had a major impact on human mental health, which has had an impact on the instability of emotions. Unstable emotions can cause stress, characterized by feelings of sadness, gloom, and unhappiness. Prolonged feelings of sadness can be a sign of depression. Therefore, tools are needed that able to provide information about the intensity of sadness felt in order to reduce prolonged sadness. Some studies have successfully created tools to detect emotions with voice signal. Therefore, the purpose of this research is to make a system that is able to detect the intensity of sad emotions through human's voice. The research was conducted using BFCC feature extraction, which has a better accuracy than the MFCC method if the sample data has a lot of noise. The system will work with the help of a microphone as a device to record. In the first process, the system will record using a microphone, the recorded results will be processed to extract the features and classify them. After the prediction results are obtained, the results will be displayed on the LCD display. This research uses the Crowd Sourced Emotional Actors Dataset (CREMA-D) consisting of several emotions with high, low, and mid levels, but in this research only focuses on the use of sad emotions. The results of the study obtained an accuracy of 60% with an average signal-to-ratio (SNR) of 23.9 dB, and has an average difference of 11.76 dB better than the MFCC method.]]>
Copyrights © 2022
