STM Journals Peer-reviewed research from Consortium e-Learning Network

Journal of Artificial Intelligence Research & Advances

Voice Recognition Technique using Guassian Mixture Model

  1. Anagha Girme
  2. Nivedita Mahato
  3. Sucheta Manve
  4. Abhijeet Banubakode

Abstract

This paper gives us some of the important information about the methods used for voice recognition system. Voice recognition is the identification of person with the help of characteristic of voice. Voice recognition has scope ranging from access control to forensics. In this system we are able to identify the speaker as well as verification of the speaker take place using feature extraction and feature matching technique using GMM. In this work the features of voice of the person are extracted with the help of Mel-Frequency Cepstral Coefficient (MFCC) and Subband based Cepstral Parameter (SBC). Hence the accuracy of speaker recognition increases giving speaker a more flexible system .In the experimental result SBC is more accurate than MFCC.Cite this ArticleAnagha Girme, Nivedita Mahato, Sucheta Manve, Abhijeet Banubakode. Voice Recognition Technique Using Guassian Mixture Model. Journal of Artificial Intelligence Research & Advances. 2015; 2(2): 12–15p.

Keywords

Support