Automated Speech Emotion Recognition App Development on Smart Phones using Cloud Computing
Journal Title: International Journal of engineering Research and Applications - Year 2018, Vol 8, Issue 5
Abstract
Speech Emotion Recognition (SER) has become a major endeavor in Human-Computer Interaction (HCI) and speech processing. Accurate SER is essential for many applications, such as assessingcustomer satisfaction with the quality of services and detecting/assessing the emotional state of children in care. The large number of studiespublished on SER reflects the demand for its use. In thispaper, the proposed system presents a novelmethod of speech recognition based on the cloud model, in combination with the traditional speech emotion system. The process of predictingemotionsfrom speech emotion audio files containsseveral stages. The first stage of this system is the pre-processing stage, whichisapplied by detecting the speech in an audio file and thenreducing the noise. The second stage involvesextractingfeaturesfrom speech emotion files using the Melfrequency cepstral coefficient (MFCC) feature extraction algorithms. This generates the training and testingdatasetsthatcontain the emotions of Anger, Disgust, Fear, Happiness, Neutral, Sadness, and Surprise. Support Vector Machine (SVM) classifiers are thenused for the classification stage in order to predict the emotion. In addition, a Confusion Matrix (CM) technique isused to evaluate the performance of theseclassifiers. The proposed system istested on SAVEEand RMLdatabases and achieved a prediction rate of 95.3%
Authors and Affiliations
Humaid Alshamsi, Veton Kepuska, Hazza Alshamisi
Keywords
Related Articles
- EP ID EP394198
- DOI 10.9790/9622-0805027177.
- Views 100
- Downloads 0
Extra Geometry I. Extra lines
Using the theory of exploded numbers by the axiom – systems of real numbers and Euclidean geometry, we introduce a the concept of extra - line of the three – dimensional space. The extra - lines are the visible subsets o...
A Novel Control Strategy For Shunt Active Power Filter Using NARX Neural Network
This paper presents a novel control strategy for shunt active power filter (APF) based on Nonlinear AutoRegressive with eXogenous neural network (NARX). The proposed series parallel NARX neural network is trained by usin...
Light Weight Concrete with Full Replacement for Fine and Partial Replacement for Coarse Aggregate
The normal concrete has a density of 2500 kg/ m 3 which makes it difficult to handle at the time of construction and increasing the sectional dimensions of structural elements. The major contributions to the density of c...
A Novel control of a Grid-Interfacing Inverter to Improve the Quality of Power at PCC for a 3-Phase 4-wire DG System
Renewable energy resources (RES) are being increas- ingly connected in distribution systems utilizing power electronic converters. This paper presents a novel control strategy for achieving maximum benefits from these gr...
Using Variational Iteration Method To Solve Burger’s Equations
The variational iteration method (VIM) is a method that is being increasingly used for solving non-linear equations. Recent applications have included using (VIM) for solving Burger’s equations. Looking at methods for re...