|
|||||
|
|
| 基于噪声被掩蔽概率的语音增强算法 | |
| 引用本文： | 王继曾,王婵飞. 基于噪声被掩蔽概率的语音增强算法[J]. 计算机仿真, 2008, 25(9) |
| 作者姓名： | 王继曾 王婵飞 |
| 作者单位： | 兰州理工大学计算机与通信学院,甘肃兰州,730050 |
| 摘 要： | 针对MMSE方法语音失真较大的缺点,提出一种将噪声被掩蔽概率引入高分辨率掩蔽感知模型的方法,通过初始噪声序列进行噪声谱的更新,然后计算噪声掩蔽参数,适时更新数据参数以动态确定每一帧的权值,实验结果表明,该方法在有效抑制背景噪声的同时还降低了音乐噪声,在语音降噪方面实现了比MMSE方法更好的增强效果. |
| 关 键 词： | 听觉掩蔽 小波变换 噪声估计 音乐噪声 语音增强 |
A Speech Enhancement Method Based on Masking Properties of Human Auditory |
|
| WANG Ji-zeng,WANG Chan-fei. A Speech Enhancement Method Based on Masking Properties of Human Auditory[J]. Computer Simulation, 2008, 25(9) | |
| Authors: | WANG Ji-zeng WANG Chan-fei |
| Affiliation: | WANG Ji-zeng,WANG Chan-fei(Department of Computer , Communication,Lanzhou University of Technology,Lanzhou Gansu 730050,China) |
| Abstract: | Abstract: In order to reduce the larger voice distortion of MMSE method,an effective method for speech enhancement is described in this paper.By utilizing an initialized noise sequence,calculating the noise masking thresholds with noise masking parameter,it can update the estimated noise instantly.Experimental result shows that the new approach can restrain noise from background reduce residual musical noise and realize better enhancement effect. |
| Keywords: | Masking of auditory Wavelet transform Noise estimate Musical noise Speech enhancement |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ | |