<![CDATA[Emotions are expressions that humans have in responding or responding to things that happen to themselves or in the environment around them, with these emotions humans are more able to express what they feel about a situation or event, these emotions can also be a means of communication other than language, because with emotions Of course, humans can know what happens to other humans around them. One of the human expressions to be able to communicate is by voice, sound can also be used to find out the type of emotion that is being experienced by the speaker. Mel-Frequency Cepstrum Coefficient (MFCC) is one of the feature extractions that is often used in the field of speech technology where in this feature extraction the human voice recording will be converted into a convolution matrix, namely a spectrogram or voice signal. K-Nearest Neighbor (K-NN) is a method that works by grouping new data based on the distance (neighborhood) from one data to the other. In the study of classical human emotions with speech using the K-Nearest Neighbor (K-NN) method, it is not appropriate to use this method because it only gets 50% accuracy.]]>
Copyrights © 2022
