High Impact Factor : 4.396 icon | Submit Manuscript Online icon |

Prosody Classification: Confirmation to Prosody Conversion

Author(s):

Priyanka Dhawale , SMt. Kashibai navale COE, Pune; Pallavi Despande, SMt. Kashibai navale COE, Pune

Keywords:

Prosody classification, prosody, prosody conversion, support vector machine, kernel function, RBF, polynomial

Abstract

Prosody classification is one of the challenging tasks in prosody conversion. To determine whether we can convert the prosody or not, prosody classifications is main task. If we can classify the prosody, then it’s sure we can convert the prosody. In this paper, we focused on the classification of prosody. We demonstrated the approach of classifying the prosody using the multi-class support vector machine. In first part, we considered spectral as well as prosodic features for classification. In second part, we considered only prosodic features. In second part, the classification accuracy is more. In SVM, we used the combination of RBF (radial basis function) and polynomial kernel function which gave better accuracy than single RBF kernel function. The successful prosody classification leads to confirmation of prosody classification.

Other Details

Paper ID: IJSRDV4I40165
Published in: Volume : 4, Issue : 4
Publication Date: 01/07/2016
Page(s): 180-182

Article Preview

Download Article