Speech Emotion Recognition using Artificial Neural Networks
Journal: International Journal of Scientific Engineering and Research (IJSER) (Vol.4, No. 5)Publication Date: 2016-05-05
Authors : P. Prithvi; T. Kishore Kumar;
Page : 8-10
Keywords : MFCC; Prosodic Features; Emotions; Neural Networks.;
Abstract
Emotion recognition in speech is a topic in which little research has been done to date. Emotion recognition in speech is an interesting and applicable research topic. In this paper, we present a system for emotion recognition using neural networks. By using a database of words, our system will be speaker independent. The classifiers will used to distinguish emotions such as neutral, anger, happy, sorrow etc. Emotional speech samples will used as database for emotion recognition from speech and extracted features from speech samples are prosodic features like pitch, energy, formats and spectral features like mel frequency cepstral coefficients for speech are used. Further the classifiers will be trained by using these features for classifying emotions accurately. Thus, many components like pre-processing of speech, MFCC computations, classifiers come together in the implementation of emotion recognition system using speech.
Other Latest Articles
- Comparative Analysis of MFCC, LFCC, RASTA-PLP
- Characterization of EIC Cellulose-Polyol Isocyanate Composite for Isolator
- Investigation on Static & Dynamic Behaviour of Stiffened Panels
- Automatic Terminal Information System (ATIS)
- Finite Element Simulation of Fillet Welded Cruciform Joint for Evaluation of Fatigue Behavior
Last modified: 2021-07-08 15:37:51