DMCA
Time Normalization of LPC Feature Using Warping Method (2005)
| Venue: | Proceedings of the International Joint Conference on Neural Networks |
| Citations: | 1 - 1 self |
BibTeX
@INPROCEEDINGS{Sudirman05timenormalization,
author = {Rubita Sudirman and Sh. Hussain Salleh and Puspa Inayat Khalid and Abd Hamid Ahmad},
title = {Time Normalization of LPC Feature Using Warping Method},
booktitle = {Proceedings of the International Joint Conference on Neural Networks},
year = {2005},
pages = {1903--1908}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
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 processed features are pitch and Linear Predictive Coefficients (LPC) for input and reference templates, based on Dynamic Time Warping (DTW) algorithm. The first task is to extract pitch features using Pitch Scale Harmonic Filter (PSHF) algorithm [12]. Another task is to align the input frames (test set) to the reference template (training set) using DTW fixing frame (DTW-FF) algorithm. This proper time normalization is needed since NN is designed to compare data of the same length whilst same speech can varies in their length. By doing frame fixing (time normalization), the test set and the training set is adjusted to the same number of frames. Having both pitch and LPC features fixed frames, speech recognition using neural network can be performed. 1.
Keyphrases
time normalization lpc feature using warping method reference template dynamic time warping speech recognition training set test set pitch feature dtw fixing frame first task reliable reference template processed feature lpc feature input feature pitch scale harmonic filter proper time normalization input frame length whilst speech linear predictive coefficient neural network artificial neural network
