DEVELOPING AN INTEGRATED MODEL USING SIGNAL AND SPEECH PROCESSING, CONTENT AND ACOUSTIC ANALYSIS TO ENHANCE THE EFFICACY OF SER USING PREDICTIVE STATISTICAL METRICS

Karan Gupta

Vol. 4, Jan-Dec 2018

Page Number: 087-101

Abstract:

Different authorities have coordinated examinations on the affirmation of feeling from human talk with different examination designs. Speech Emotion Recognition (SER) is a specific class of signal processing where the principal objective is to recognize the energetic state of people from voice. Effects of acoustic parameters, the authenticity of the data used, and execution of the classifiers have been the pivotal issues for feeling affirmation investigate the field. 81 s (distributed in the listed diaries) have been assessed by the approaches used for feeling stamping, acoustic features and classifiers and the database used. The principle point is to examine: portray the highlights of the databases being used and to make a brief on the proficiency of acoustic parameters and the classifiers utilized by the past examinations.

Back Download