| 融合SS、MFCC和PMC技术的语音去噪方法 |
| |
| 引用本文： | 丁冬冬,佘玉梅,江涛,庄丽,王米利,刘敬凤. 融合SS、MFCC和PMC技术的语音去噪方法[J]. 云南民族大学学报(自然科学版), 2014, 0(3): 232-234 |
| |
| 作者姓名： | 丁冬冬 佘玉梅 江涛 庄丽 王米利 刘敬凤 |
| |
| 作者单位： | 云南民族大学数学与计算机科学学院,云南昆明650031 |
| |
| 基金项目： | 国家自然科学基金（61363022）;云南省自然科学基金（2011FZ176）;云南省计算机科学与技术专业核心课程教学团队项目. |
| |
| 摘 要： | 为提高语音识别系统在噪音情况下的识别率,提出了一种融合信号级去噪、参数级去噪、模型级去噪的方法.首先用谱减法对带噪的语音信号进行去噪,再利用Mel倒谱系数(MFCC)对处理后的语音信号进行特征提取,最后经过并行模型结合处理法(PMC)处理得到较高识别率的语音信号.
|
| 关 键 词： | 语音信号 谱减法 MFCC PMC 信噪比 |
A denosing method based on an integration of SS,MFCC and PMC technology |
| |
| DING Dong-dong,SHE Yu-mei,JIANG Tao,ZHUANG Li,WANG Mi-li,LIU Jing-feng. A denosing method based on an integration of SS,MFCC and PMC technology[J]. Journal of Yunnan Nationalities University:Natural Sciences Edition, 2014, 0(3): 232-234 |
| |
| Authors: | DING Dong-dong SHE Yu-mei JIANG Tao ZHUANG Li WANG Mi-li LIU Jing-feng |
| |
| Affiliation: | (School of Mathematics and Computer Science, Yunnan University of Nationalities, Kunming 650031, China) |
| |
| Abstract: | In order to improve the recognition rate of the speech sound system under noisy circumstances, one special method is put forward in this paper, whichintegrates the methods of the signal denoising, parameter denoising and model denoising. This method first employs spectral subtraction to denoise the system and then extracts the features with the aid of MFCC. Finally, after PMCC -processing, the speech sound signal of a higher recognition rate is attained. |
| |
| Keywords: | speech signal spectral subtraction MFCC PMC signal - to - noise rate |
| 本文献已被 维普 等数据库收录！ |
|
