<![CDATA[Crying is a baby's way of communicating. The sound of crying can be used to identify problems in babies, such as hunger, pain, drowsiness, fatigue, discomfort, coldness or heat, and others. However, not everyone can recognize the meaning of the crying baby. The combination of MFCC and DWT features was used in this study as an extraction feature in baby crying sounds. In this study, the Convolutional Neural Network (CNN) method is used to classify the meanings of sounds from babies crying. The dataset used in this study is a public dataset consisting of a total of 61 training data and 30 testing data. The types of crying babies used in this study were hunger, fatigue, discomfort, and stomach ache. Based on the test results, the MFCC and CNN features obtained a precision of 32.76%, a recall of 32.63%, and an accuracy of 73.33%. The combination of MFCC and DWT features (Mean, Standard Deviation, Range, Max) and CNN obtained a precision of 50.91%, a recall of 44.23%, and an accuracy of 73.33%.]]>
Copyrights © 2021
