CNN based approach for Speech Emotion Recognition Using MFCC, Croma and STFT Hand-crafted features

Abstract

At present, speech emotion recognition (SER) is a very challenging and demanding research area because of its wide real-life applications. For SER there are two major challenges: first one is to identify the relevant feature vector and the second one is to identify the suitable classifier. To overcome these challenges, we proposed a model for SER. In this approach firstly we extract the hand-crafted features Mel frequency cepstral coefficient (MFCC), Croma and Short-term Fourier Transform (STFT) from the emotional speech signals and these extracted features are considered as input for the considered deep learning classifying technique, Convolutional Neural Networks (CNN). This work investigates the credibility of convolutional neural networks using hand-crafted features MFCC with value 13, Croma with value 13 and STFT. The proposed model comprises 8 CNN layers and output of the last layer is faded to flatten layer and then apply soft-max activation function to identify the emotions. We employed the publicly accessible speech emotional databases RAVDESS, TESS, and SAVEE, as well as their combinations to evaluate the performance of our proposed model. Experiments show that in terms of average accuracy, the suggested model outperforms the current state-of-the-art SER techniques.