|Table of Contents|

Feature Extraction and Speech Recognition Based on An Auditory Model(PDF)

《南京理工大学学报》(自然科学版)[ISSN:1005-9830/CN:32-1397/N]

Issue:
1998年02期
Page:
21-24
Research Field:
Publishing date:

Info

Title:
Feature Extraction and Speech Recognition Based on An Auditory Model
Author(s):
ZhangYan ZhangJie HuangZhitong
School of Information, NUST, Nanjing 210094
Keywords:
phonet ic recog nit ion signal processing audito ry model
PACS:
TN912.34
DOI:
-
Abstract:
T his paper analyses the principle of speech sig nal pro cessing of human co chlea and audito ry ner ve. The fo ur stag es of pr ocessing include cochleaqs f requency analysis of sound, Organ of Cor tiqs ex chang ing f rom mechanical energ y to chemical ener gy , lateral inhibitory netw ork ( LIN) and g ain adjust ing . At every stage, mathemat ics models are given. The o utput of the last model is the featur e used for speech r ecognit io n ( SR) . The result of SR test show s that the feature ex t racted f rom this audito ry model r epresent s speech sig nal ex act ly .

References:

1 杨行峻, 迟惠生. 语音信号数字处理. 北京: 电子工业出版社, 1995, 34~46
2 Shamma S A, Chadw ick R, Wilbur J. A Bio phy sical Mo del of Cochlea Pr o cessing : Intensity Dependence o f Pure T one Responses. J. Acoust. So c. Am, 1986, 80( 1) : 133~145
3 Yang Xiaow ei, Wang Kuansan, Samma S A. Auditor y Repr esentat ions of Aco ustic Sig nals. IEEE T ransactio ns o n Informatio n Theor y , 1992, 38( 2) : 824~ 839
4 胡守仁, 余少波, 戴葵. 神经网络导论. 长沙: 国防科技大学出版社, 1992, 201~205
5 林宝成, 黄志同. 基于听觉模型的子波变换语音处理. 数据采集与处理, 1995, 10( 4) : 269~274
6 张焱, 张杰, 张生平等. 基于听觉模型的鲁棒性语音识别. 见: 王炳锡. 语音图像与通信号处理第八届全国学术会议论文集. 郑州: 解放军信息工程学院, 1997. 106~109

Memo

Memo:
-
Last Update: 2013-03-29