Enhancing the accuracy in classifying human emotion via speech recognition using novel support vector machine compared with convolution neural network

M. Naren, M. Sandhiya

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

Abstract

Speech is the most prevalent way that people communicate with one another. Based on various methods for classifying the emotions from transformed data, communication systems identify the emotional states of people. Voice and visual media are the most direct information delivery methods used by humans. MFCC is the most often utilized extraction feature recognition. Speech data preprocessing, feature extraction, and speech emotion classification make up the three stages of the speech emotion detection process. Data may occasionally show abnormalities. As a result, the key of emotion recognition lies in the categorization architecture and speech emotion characteristic, both of which contain crucial knowledge.

Original languageEnglish
Title of host publicationInternational Conference on Information Technology and Mechatronics Engineering, ICITME 2021
EditorsDexter R. Buted, Elbert M. Galas, Randy Joy M. Ventayen, Potenciano D. Conte
PublisherAmerican Institute of Physics Inc.
ISBN (Electronic)9780735444904
DOIs
StatePublished - 16 May 2023
Externally publishedYes
Event2021 International Conference on Information Technology and Mechatronics Engineering, ICITME 2021 - Pangasinan, Virtual, Philippines
Duration: 10 Dec 202112 Dec 2021

Publication series

NameAIP Conference Proceedings
Volume2602
ISSN (Print)0094-243X
ISSN (Electronic)1551-7616

Conference

Conference2021 International Conference on Information Technology and Mechatronics Engineering, ICITME 2021
Country/TerritoryPhilippines
CityPangasinan, Virtual
Period10/12/2112/12/21

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 convolution neural network'. Together they form a unique fingerprint.

Cite this