Research Article
Speech Emotion Recognition using Multilayer Perceptron Classifier on Ravdess Dataset
@INPROCEEDINGS{10.4108/eai.7-12-2021.2314726, author={Rekha R and Tharani R S}, title={Speech Emotion Recognition using Multilayer Perceptron Classifier on Ravdess Dataset}, proceedings={Proceedings of the First International Conference on Combinatorial and Optimization, ICCAP 2021, December 7-8 2021, Chennai, India}, publisher={EAI}, proceedings_a={ICCAP}, year={2021}, month={12}, keywords={ser mfcc chroma mel mlp}, doi={10.4108/eai.7-12-2021.2314726} }- Rekha R
Tharani R S
Year: 2021
Speech Emotion Recognition using Multilayer Perceptron Classifier on Ravdess Dataset
ICCAP
EAI
DOI: 10.4108/eai.7-12-2021.2314726
Abstract
Emotion is the best way to express one person’s thought and action to others. Identifying emotions from one person’s speech is the most required technology for today’s world. Emotion Recognition can be much helpful to derive various useful insights about the thoughts of a person. Speech emotion recognition (SER) is the process of obtaining emotions like happiness, sadness, neutral and other emotions from one person’s speech. In this paper, Speech emotion recognition method is used to gain emotions from RAVDESS dataset. The emotion extraction is done based on speech features like Mel-frequency cepstrum coefficients (MFCC), chroma and mel. The extracted data is then trained with Multilayer perceptron classifier (MLP) and obtained an accuracy of 82.8%..