一种新的抗噪语音特征的提取

黄亚娟, 郑林. 一种新的抗噪语音特征的提取[J]. 微电子学与计算机, 2011, 28(10): 215-216,220.

引用本文:

黄亚娟, 郑林. 一种新的抗噪语音特征的提取[J]. 微电子学与计算机, 2011, 28(10): 215-216,220.

HUANG Ya-juan, ZHENG Lin. A New Antinoise Feature Extraction of Speech[J]. Microelectronics & Computer, 2011, 28(10): 215-216,220.

Citation:

HUANG Ya-juan, ZHENG Lin. A New Antinoise Feature Extraction of Speech[J]. Microelectronics & Computer, 2011, 28(10): 215-216,220.

A New Antinoise Feature Extraction of Speech

摘要: 针对LPCC易受噪声干扰和不能反映人耳听觉特性的缺点, 提出了新的抗噪声特征, 实验表明, 在各种信噪比 (SNR) 环境中, 该方法的精度高于线性预测倒谱系数 (LPCC) 和美尔倒谱系数 (MFCC) .

Abstract: As LPCC is easily disturbed by noise and it doesn't contain human auditory characteristics, Proposing a new antinoise feature.The experiment showed that in all kinds of SNR environment, the method has more accuracy than LPCC and MFCC.

/