Isolated Spoken Odia Digit Recognition Using Support Vector Machine

Over the last few decades, significant fur-therance is being ascertained on speech recognition technology, devising its approach from test bed to real prospect. An automatic digit recognition system always helpful for physically challenged persons (blind people) or elder people to have a telephonic...

Full description

Saved in:
Bibliographic Details
Published in2019 International Conference on Applied Machine Learning (ICAML) pp. 147 - 152
Main Authors Mohanty, Prithviraj, Nayak, Ajit Kumar
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.05.2019
Subjects
Kernel Mapping
Machine learning
MFCC
RBF
Sigmoid
SVM
Online AccessGet full text

Cover

More Information
Summary:Over the last few decades, significant fur-therance is being ascertained on speech recognition technology, devising its approach from test bed to real prospect. An automatic digit recognition system always helpful for physically challenged persons (blind people) or elder people to have a telephonic conversion, setting the pin number for their debit and credit cards and also having the security code for some applications without physically touching the system. The work presented on this paper emphasize the recognition of isolated Odia digits using Support Vector Machine (SVM). The feature parameters for the spoken digits are extracted using Mel-Frequency Cepstral Coefficient (MFCC) and provided as the inputs to the recognition process. Different kernel mapping functions such as: polynomial, sigmoid and Radial Basis Function (RBF) are used in order to construct the SVM classification models. The experiments are evaluated and the result obtained withdraws a better accuracy compared to the earlier one which uses HMM model for isolated Odia digit recognition.
DOI:10.1109/ICAML48257.2019.00036