Speech Emotion Recognition Based on Multiple Acoustic Features and Deep Convolutional Neural Network

Speech emotion recognition (SER) plays a vital role in human–machine interaction. A large number of SER schemes have been anticipated over the last decade. However, performance systems is challenging due to high complexity systems, poor feature distinctiveness, and noise. This paper presents acoustic set based on Mel frequency cepstral coefficients (MFCC), linear prediction (LPCC), wavelet packet transform (WPT), zero crossing rate (ZCR), spectrum centroid, spectral roll-off, kurtosis, root mean square (RMS), pitch, jitter, shimmer improve distinctiveness. Further, lightweight compact one-dimensional deep convolutional neural network (1-D DCNN) used minimize computational represent long-term dependencies speech signal. The overall effectiveness proposed systems’ evaluated Berlin Database Emotional (EMODB) Ryerson Audio-Visual Song (RAVDESS) datasets. system gives an accuracy 93.31% 94.18% for EMODB RAVDESS datasets, respectively. MFCC 1-D DCNN provide greater outpace traditional techniques.