An Examination of Emotion Recognition using Machine Learning Algorithms on Different Speech Databases
Kogila Raghu1, M Sadanandam2, V Kamakshi Prasad3
1K Raghu, Research Scholar, Department of Computer Science & Engineering, Kakatiya University, Warangal (Telangana), India.
2Dr. M Sadanandam, Assistant Professor, Department of Computer Science & Engineering, Kakatiya University, Warangal (Telangana), India.
3V Kamakshi Prasad, Professor, Department of Computer Science and Engineering, JNTUH College of Engineering Hyderabad, JNT University, Hyderabad (Telangana), India.
Manuscript received on 24 February 2020 | Revised Manuscript received on 04 March 2020 | Manuscript Published on 15 March 2020 | PP: 37-39 | Volume-9 Issue-4S2 March 2020 | Retrieval Number: D10090394S220/2020©BEIESP | DOI: 10.35940/ijitee.D1009.0394S220
Open Access | Editorial and Publishing Policies | Cite | Zenodo | Indexing and Abstracting
© 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: The speech recognition system plays a vital role in understanding the emotions of natural language. The identification of emotions from speech is a challenging task. The performance of the speech recognition system is effects on the speech signals. The speech contains different emotions feelings. Many researchers introduced different emotion recognition techniques. However, these techniques achieved better performance but unsatisfied in identify emotion of natural languages. This paper proposed a novel speech recognition system, which identify the emotions based on the speech signals.The Mel Frequency Cepstral Coefficients (MFCC) features. On the resultant features of speech applied crossvalidation using the test emotions. The performance of the proposed system verify with the SVM and other two classifiers. The proposed emotion recognition system achieves better performance. The empirical results shows that the proposed system outperforms when compare with different classifiers and databases.
Keywords: Corpora, Features LPCC, MFCCLR, SVM, HMM, GMM.
Scope of the Article: Machine Learning