An Improved Hindi Speech Emotion Recognition System
Agnes Jacob1, P. Mythili2
1Agnes Jacob, Research Scholar, Department of Electronics, Engineering, Cochin University of Science and Technology, Kochi (Kerala), India.
2Dr. P. Mythili, Head, Department of Electronics, Engineering, Cochin University of Science and Technology, Kochi (Kerala), India.
Manuscript received on 10 November 2013 | Revised Manuscript received on 18 November 2013 | Manuscript Published on 30 November 2013 | PP: 25-29 | Volume-3 Issue-6, November 2013 | Retrieval Number: F1320113613/13©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | 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: This paper presents the results of investigations in speech emotion recognition in Hindi, using only the first four formants and their bandwidths. This research work was done on female speech data base of nearly 1600 utterances comprising neutral, happiness, surprise, anger, sadness, fear and disgust as the elicited emotions. The best of the statistically preprocessed formant and bandwidth features were first identified by the KMeans, K nearest Neighbour and Naive Bayes classification of individual features. This was followed by artificial neural network classification based on the combination of the best formants and bandwidths. The highest overall emotion recognition accuracy obtained by the ANN method was 97.14%, based on the first four values of formants and bandwidths. A striking increase in the recognition accuracy was observed when the number of emotion classes was reduced from seven. The obtained results presented in this paper, have not been reported so far for Hindi, using the proposed spectral features as well as with the adopted preprocessing and classification methods.
Keywords: Formant, Emotion, Kmeans, K Nearest Neighbour, Naive Bayes, Artificial Neural Network.
Scope of the Article: Pattern Recognition