Infant Cry Analysis

Abstract: In this research, we have proposed a machine learning model that works on Random Forest Classifier, which extracts the MFCCs(Mel-Frequency Cepstral Coefficients) from baby cries and utilizes these features for predictions such as hungry, belly-pain, burping, tired and discomfort. This research can help the parents, caregivers to determine the exact reason behind the crying baby and suggesting the necessary actions to be taken further depending upon the baby cry.

Keywords: MFCCs(Mel-Frequency Cepstral Coefficients), FFT(Fast Fourier Transform), ML(Machine Learning), DL(Deep Learning), LSTM(Long Short Term Memory).