ResearchBib Share Your Research, Maximize Your Social Impacts
Sign for Notice Everyday Sign up >> Login


Journal: International Journal of Advanced Research in Engineering and Technology (IJARET) (Vol.11, No. 12)

Publication Date:

Authors : ; ;

Page : 2071-2079

Keywords : Speech; Speech Emotion Recognition; Machine Learning; Mel-frequency Cepstral Coefficient; Gaussian Mixture Model.;

Source : Downloadexternal Find it from : Google Scholarexternal


Machine learning is a mechanism to perform interaction between human and machine and also to provide smart and sustainable computing. Speech is a most attractive and effective way for expressing emotion as well as attitude. This paper focus on identifying impact of gender on different basic emotions during exchange of speech. To analyze above different emotional features, emotion speech Hindi database simulated by Indian Institute of Technology Kharagpur, Mel-frequency Cepstral Coefficient feature extraction method and a classification method are processed. The analysis shows 75.46% b, 71.59% and 68.06% for 32-centered, 16-centered and 8- centered respectively when it is tested with text dependent gender specific(Female) data whereas 74.90%, 71.11% and 65.89% for 32-centered, 16-centered and 8- centered respectively for text dependent gender specific(Male) data. Simulation also carried out for text independent data. The simulation is carried out by using Indian Institute of Technology Speech emotion for Hindi database. Simulation clearly shows the recognition always happens good when it is performed by female speech than male. And also it doesn't matter, whether it is text dependent or text independent.

Last modified: 2021-02-24 18:05:08