| 加性噪声环境下语音特征参数提取方法的研究 |
| |
| 引用本文： | 张昊慧.加性噪声环境下语音特征参数提取方法的研究[J].淮阴师范学院学报(自然科学版),2010,9(4). |
| |
| 作者姓名： | 张昊慧 |
| |
| 作者单位： | 淮阴师范学院,物理与电子电气工程学院,江苏,淮安,223300;东南大学,信息科学与工程学院,江苏,南京,210096 |
| |
| 摘 要： | 提出一种具有良好抗噪性的语音特征分析方法.将语音信号的短时自相关序列进行时间方向上的平滑处理,然后利用平滑后的序列代替原语音信号进行线性预测分析,从而得到线性预测倒谱系数.实验表明,利用该特征参数的语音识别系统的识别性能优于MEL倒谱系数、LPC倒谱系数等传统的语音特征参数.
|
| 关 键 词： | 短时自相关函数 平滑处理 LPCC |
A Study of Speech Features Extraction under Noisy Conditions |
| |
| ZHANG Hao-hui.A Study of Speech Features Extraction under Noisy Conditions[J].Journal of Huaiyin Teachers College(Natrual Science Edition),2010,9(4). |
| |
| Authors: | ZHANG Hao-hui |
| |
| Institution: | ZHANG Hao-hui1,2(1.School of Physics and Electronic Electrical Engineering,Huaiyin Normal University,Huaian Jiangshu 223300,China)(2.School of Information Science and Engineering,Southeast University,Nanjing 210096,China) |
| |
| Abstract: | A good anti-noise characteristics of speech analysis is introduced.The idea is to filter the short-time Autocorrelation Sequence of speech.Then frequency warped LPC algorithm is applied to the Sequence instead of the original speech.The recognition experiment shows the feature is more effective than MEL Cepstrum and LPC cepstrum coefficients. |
| |
| Keywords: | LPCC |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
