King Saud University Repository >
King Saud University >
Science Colleges >
College of Computer and Information Sciences >
College of Computer and Information Sciences >

Please use this identifier to cite or link to this item: http://hdl.handle.net/123456789/14962

Title: Effect of articulatory trajectories on phoneme recognition performance
Authors: Mohammad Mahedi Hasan
Ghulam Muhammad
Keywords: articulatory features, hidden markov model, Gram-Schmidt orthogonalization, multilayer neural networks
Issue Date: 2010
Publisher: IEEE Computer Society
Abstract: This paper presents a method for describing the effect of articulatory trajectories on phoneme recognition. The proposed method comprises three stages. The first stage embeds three multilayer neural networks (MLNs): MLN (LF-DPF) that maps acoustic features or local features (LFs) onto articulatory features or distinctive phonetic features (DPFs), MLN (cntxt) that reduces misclassifications at phoneme boundaries, and MLN (Dyn) that controls dynamics of DPF features. The second stage incorporates an inhibition/enhancement (In/En) network by varying the trajectories of articulators to achieve categorical DPF movement by enhancing DPF peak patterns and inhibiting DPF dip patterns. The third stage decor relates continuous DPF vectors using the Gram-Schmidt algorithm before feeding into a hidden Markov model (HMM)-based classifier. In the experiments on Japanese Newspaper Article Sentences (JNAS) database, the proposed feature extractor shows the performance of phoneme recognition with the variation of different trajectories.
URI: http://hdl.handle.net/123456789/14962
Appears in Collections:College of Computer and Information Sciences

Files in This Item:

File Description SizeFormat
Dr Gulam-28-conf.docx14.82 kBMicrosoft Word XMLView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.


DSpace Software Copyright © 2002-2009 MIT and Hewlett-Packard - Feedback