|
Title:
|
SPEAKER IDENTIFICATION SYSTEM USING LIFTING WAVELET FILTERS |
|
Author(s):
|
Shingo Fukata, Shigeru Takano, Yoshihiro Okada, Kiyotaka Fujisaki |
|
ISBN:
|
978-972-8939-46-5 |
|
Editors:
|
Piet Kommers and Pedro Isaías |
|
Year:
|
2011 |
|
Edition:
|
Single |
|
Keywords:
|
Speaker identification, Lifting Wavelet Filters, LPC Cepstrum, Genetic Algorithm |
|
Type:
|
Poster/Demonstration |
|
First Page:
|
583 |
|
Last Page:
|
586 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
This paper develops a speaker identification system using lifting wavelet filters. Our aim is to design the lifting wavelet filters so as to extract the personal features efficiently. By applying the lifting wavelet filters to voice signals, low-pass and high-pass components are obtained. The personal features are extracted from the obtained low-pass components. Here the low-pass components can be tuned by free parameters included in the lifting wavelet filters. Using these parameters, we construct the new filters so as to increase the accuracy of our speaker identification system. In experiment, these parameters are determined by genetic algorithm, and we show that our system outperforms the conventional speaker identifications. |
|
|
|
|
|
|
