A comparative Analysis of ML & DL Techniques for Speech Emotion Recognition

Abstract

We pointed toward learning profound feeling elements to perceive speech emotion. Element withdrawal, Trait determination and grouped into three primary phases of the feeling acknowledgment. The principle point of this work is to further develop the discourse feeling acknowledgment pace of a framework utilizing the diverse component draw out calculations. The work accentuates on the preprocessing of the got sound examples where the clamor from discourse tests is eliminated utilizing channels. In subsequent stage, Mel Frequency Cepstral Coefficients (MFCC), Discrete Wavelet Transform (DWT), pitch, energy and Zero crossing point rate (ZCR) calculations are utilized in separating elements. While including determination stage worldwide element calculation is utilized to eliminate excess data from highlights and to recognize feelings from extricated highlights AI characterization calculations are utilized. These elementsdraw out calculations are approved for widespread feelings containing Anger, Happiness, Sad and Neutral.