Emotion Recognition from Chhattisgarhi Speech using Neural Network
Yachana Gaikwad1, Yogesh Rathore2
1Yachana Gikwad, Department of Computer Science & Technology, C.S.V.T.U. RIT College, Raipur, India.
2Yogesh Rathore, Department of Computer Science, C.S.V.T.U. RIT College, Raipur, India.
Manuscript received on November 22, 2014. | Revised Manuscript received on December 03, 2014. | Manuscript published on December 30, 2014. | PP: 153-158 | Volume-4 Issue-2, December 2014. | Retrieval Number: B3656124214/2013©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Speech Emotion Recognition (SER) is emerging as a crucial research area. Many works have been done in field of SER for example speaker dependent / independent SER system, language dependent/independent SER system, extracting different emotions like happiness, anger, sadness, disgust, boredom, neutral. All these works have been achieved by working on speech sample, for which we need speech emotion database. In this work we have developed speech emotion database in Chhattisgarhi language. Speech emotions can be recognized by using different features of speech, which may be prosodic feature (pitch, energy) or phonetic features (MFCC, Format Frequency) etc. Thus for selecting feature of speech for emotion identification, a review of works on speech emotion recognition is given in this paper. The aim of this paper is to present the works which are important to design and develop SER system for Chhattisgarhi language using neural network and analyze it.
Keywords: Energy, Format Frequency, MFCC, Pitch, Neural Network.