[1]马明,张焱,王建宇,等.语音信号基本载体的研究[J].南京理工大学学报(自然科学版),1996,(05):57-60.
 MaMing ZhangYan WangJianyu HuangZhitong.A Study on the Essential Feature of Speech Signals[J].Journal of Nanjing University of Science and Technology,1996,(05):57-60.
点击复制

语音信号基本载体的研究()
分享到:

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

卷:
期数:
1996年05期
页码:
57-60
栏目:
出版日期:
1996-10-30

文章信息/Info

Title:
A Study on the Essential Feature of Speech Signals
作者:
马明张焱王建宇黄志同
南京理工大学信息学院, 南京210094
Author(s):
MaMing ZhangYan WangJianyu HuangZhitong
School of Information,NUST, Nanjing 210094
关键词:
语音信号处理 语音识别 数据压缩
Keywords:
speech sing al processing speech r ecognit io n data compressing
分类号:
TP391.42
摘要:
语音信号特征的提取是语音识别的基础。目前使用的特征提取方法要将语音信号经多次复杂变换(如傅里叶变换、自相关变换等)后提取特征。这样,往往满足不了实时性要求。该文对原始语音信号做简单处理(如切顶、转换成直方波形等)的实验,证明经简单处理后语音信号波形比原始波形简单、直观,计算频域特征和自相关特征简单方便、实时性好,并有望从中进一步发现新的语音特征提取方法
Abstract:
It is the most impor tant step in speech recognitio n to ex tr act the essent ial featur es of speech signal s. Now , the w idely used methods of essential features ex tr act ing are to tr ansform the speech sig nals complexily many t imes, such as Fo urier tr ansform, auto- cor relat ion tr ansform. T hese can not meet the needs of r eal-t ime acquirement . Authors make a lo t of ex periment s of speech sig nals simple tr ansforming , such as cut t ing-top, rectang ular w ave. It show s that the t ransformed speech sig nals can be recog nized by humans, and the t ransformed speech w aves are simpler than the original speech w aves in pro cessing . It is ho peful to f ind a new method of speech features ex tr act ing .

参考文献/References:

1 胡光锐. 语言处理与识别. 上海: 科学技术文献出版社, 1994
2 陈尚勤, 罗承烈, 杨雪. 近代语音识别. 成都: 电子科技大学出版社, 1991

相似文献/References:

[1]路建伟,丁庆海,朱雪平,等.战场环境下的军事命令识别技术[J].南京理工大学学报(自然科学版),2002,(04):438.
 LuJianwei DingQinghai ZhuXueping XiongGang.The Military Order Recognition Technology in Battlefield Environment[J].Journal of Nanjing University of Science and Technology,2002,(05):438.
[2]林宝成,黄志同.用于语音识别中的SOFM矢量量化方法[J].南京理工大学学报(自然科学版),1996,(01):59.
 Lin Baoeheng,Huing Zhitong.A Method of SOFM as a Vector Quantizer in Speech Recognition[J].Journal of Nanjing University of Science and Technology,1996,(05):59.

备注/Memo

备注/Memo:
江苏省自然科学基金,南京理工大学科研发展基金
马 明 男 28 岁 博士生
更新日期/Last Update: 2013-04-11