Pre-processing of input features using LPC and warping process

This paper presents pre-processing of input features to artificial neural network (NN). This is for preparation of reliable reference templates for the set of words to be recognized. The first task is to extract pitch features using Pitch Scale Harmonic Filter (PSHF) algorithm. Another tas...

Full description

Bibliographic Details
Main Authors: Sudirman, Rubita, Sh-Hussain, Salleh, Ming, Ting Chee
Format: Article
Language:English
Published: 2005
Subjects:
Online Access:http://eprints.utm.my/1574/
http://eprints.utm.my/1574/1/ccsp1.pdf
_version_ 1848890166422274048
author Sudirman, Rubita
Sh-Hussain, Salleh
Ming, Ting Chee
author_facet Sudirman, Rubita
Sh-Hussain, Salleh
Ming, Ting Chee
author_sort Sudirman, Rubita
building UTeM Institutional Repository
collection Online Access
description This paper presents pre-processing of input features to artificial neural network (NN). This is for preparation of reliable reference templates for the set of words to be recognized. The first task is to extract pitch features using Pitch Scale Harmonic Filter (PSHF) algorithm. Another task is to align the input frames (test set) to the reference template (training set) using a modified DTW algorithm called DTW fixing frame (DTW-FF)algorithm. This proper time normalization is needed since NN is designed to compare data of the same length; same speech can varies in their duration. By performing frame fixing or time normalization, the test set and the training set is adjusted to a fix number of frames throughout the sets utilizing the local distance score of the matched features. Then those features can be adapted to NN for further recognition tuning.
first_indexed 2025-11-15T20:37:45Z
format Article
id utm-1574
institution Universiti Teknologi Malaysia
institution_category Local University
language English
last_indexed 2025-11-15T20:37:45Z
publishDate 2005
recordtype eprints
repository_type Digital Repository
spelling utm-15742012-01-05T02:18:19Z http://eprints.utm.my/1574/ Pre-processing of input features using LPC and warping process Sudirman, Rubita Sh-Hussain, Salleh Ming, Ting Chee TK Electrical engineering. Electronics Nuclear engineering This paper presents pre-processing of input features to artificial neural network (NN). This is for preparation of reliable reference templates for the set of words to be recognized. The first task is to extract pitch features using Pitch Scale Harmonic Filter (PSHF) algorithm. Another task is to align the input frames (test set) to the reference template (training set) using a modified DTW algorithm called DTW fixing frame (DTW-FF)algorithm. This proper time normalization is needed since NN is designed to compare data of the same length; same speech can varies in their duration. By performing frame fixing or time normalization, the test set and the training set is adjusted to a fix number of frames throughout the sets utilizing the local distance score of the matched features. Then those features can be adapted to NN for further recognition tuning. 2005-11-14 Article NonPeerReviewed application/pdf en http://eprints.utm.my/1574/1/ccsp1.pdf Sudirman, Rubita and Sh-Hussain, Salleh and Ming, Ting Chee (2005) Pre-processing of input features using LPC and warping process. 1st Conference on Computers, Commmunications, and Signal Processing . pp. 300-303. ISSN 1-4244-0012-0
spellingShingle TK Electrical engineering. Electronics Nuclear engineering
Sudirman, Rubita
Sh-Hussain, Salleh
Ming, Ting Chee
Pre-processing of input features using LPC and warping process
title Pre-processing of input features using LPC and warping process
title_full Pre-processing of input features using LPC and warping process
title_fullStr Pre-processing of input features using LPC and warping process
title_full_unstemmed Pre-processing of input features using LPC and warping process
title_short Pre-processing of input features using LPC and warping process
title_sort pre-processing of input features using lpc and warping process
topic TK Electrical engineering. Electronics Nuclear engineering
url http://eprints.utm.my/1574/
http://eprints.utm.my/1574/1/ccsp1.pdf