Hierachical Arabic Phoneme Recognition using MFCC Analysis

Hierachical Arabic Phoneme Recognition using MFCC Analysis

In this paper, a hierarchical Arabic phoneme recognition system is proposed in which Mel FrequencyCepstrum Coefficients (MFCC) features is used to train the hierarchical neural networks architecture.Here, separate neural networks (subnetworks) are to be recursively trained to recognize subsets ofpho...

Full description

Bibliographic Details
Main Authors: Abduladhem Abdulkareem Ali, I. T. Hwaidy
Format: Article
Language:English
Published: College of Engineering, University of Basrah 2007-01-01
Series:Iraqi Journal for Electrical and Electronic Engineering
Subjects:
speech recognition
phoneme recognition
hierarchical systems
Mel cepstrum
neural networks
Online Access:http://ijeee.org/volums/volume3/IJEEE3PDF/39.pdf

Internet

http://ijeee.org/volums/volume3/IJEEE3PDF/39.pdf

Similar Items