Enhancing the Accuracy in Classifying Human Emotion via Speech Recognition Using Novel Support Vector Machine Compared With Multi-layer Perceptron Classifier

M. Naren, M. Sandhiya

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Support Vector Machine and Multi-Layer Perceptron Classifier were used to improve the accuracy of identifying human speech based on emotions. The goal of the paper is to improve the accuracy of speech emotion classification. Support Vector Machine (SVM) and Multi-Layer Perceptron Classifier are the two groups in this study (MLPC). Each group has a sample size of ten people, and the study parameters are alpha = 0.05, beta = 0.2, and G power = 0.8. Their accuracies are also compared to one another using different sample sizes. With p =0.202, the Support Vector Machine is 81.8 percent more accurate than the Multi-Layer Perceptron Classifier in classifying human voice emotion. The SVM model outperforms the MLPC when it comes to detecting human emotion through speech. It can also be thought of as a superior choice for categorizing speech emotion.

Original languageEnglish
Title of host publicationAIP Conference Proceedings
EditorsAbhishek Dasore, Upendra Rajak, Manoj Panchal, Bukke Kiran Naik, Konijeti RamaKrishna
PublisherAmerican Institute of Physics Inc.
Edition1
ISBN (Electronic)9780735447356
DOIs
StatePublished - 21 Nov 2023
Externally publishedYes
Event1st International Conference on Contemporary Innovations in Mechanical Engineering, CIME 2022 - Nandyal City, India
Duration: 22 Apr 202223 Apr 2022

Publication series

NameAIP Conference Proceedings
Number1
Volume2821
ISSN (Print)0094-243X
ISSN (Electronic)1551-7616

Conference

Conference1st International Conference on Contemporary Innovations in Mechanical Engineering, CIME 2022
Country/TerritoryIndia
CityNandyal City
Period22/04/2223/04/22

Keywords

  • Emotion Classification
  • Feature Extraction
  • Multi-Layer Perceptron
  • Novel Support Vector Machine
  • Speech Emotion Detection

ASJC Scopus subject areas

  • General Physics and Astronomy

Fingerprint

Dive into the research topics of 'Enhancing the Accuracy in Classifying Human Emotion via Speech Recognition Using Novel Support Vector Machine Compared With Multi-layer Perceptron Classifier'. Together they form a unique fingerprint.

Cite this