| 失真控制下的短时谱估计语音增强算法 |
| |
| 引用本文： | 刘晓明,班超帆,冯晓荣.失真控制下的短时谱估计语音增强算法[J].西安交通大学学报,2011,45(8):78-84. |
| |
| 作者姓名： | 刘晓明 班超帆 冯晓荣 |
| |
| 作者单位： | 重庆大学通信工程学院,400030,重庆 |
| |
| 基金项目： | 国家自然科学基金资助项目(51035008) |
| |
| 摘 要： | 针对传统谱估计增强算法易产生语音畸变、导致语音清晰度低的问题,提出了一种失真控制下的短时谱估计语音增强的新算法.该算法首先引入语音畸变的客观度量参数,并根据这一参数得到抑制语音畸变的约束条件,然后结合人耳听觉掩蔽特性和无语音概率参数,修正最小均方误差对数谱估计函数,最后联立约束条件和估计函数,得到增强后的语音,从而实现了在噪声抑制和语音畸变之间的折中,改善了语音增强的效果.主观试听和客观测试结果均表明,与其他谱减法相比,在相同的信噪比和去噪度条件下,新算法的语音畸变度最小且几乎察觉不到音乐噪声.
|
| 关 键 词： | 语音增强 噪声抑制 语音清晰度 失真控制 |
A Short-Time Spectrum Estimation Algorithm of Speech Enhancement under the Distortion Control |
| |
| LIU Xiaoming,BAN Chaofan,FENG Xiaorong.A Short-Time Spectrum Estimation Algorithm of Speech Enhancement under the Distortion Control[J].Journal of Xi'an Jiaotong University,2011,45(8):78-84. |
| |
| Authors: | LIU Xiaoming BAN Chaofan FENG Xiaorong |
| |
| Institution: | LIU Xiaoming,BAN Chaofan,FENG Xiaorong(College of Communication Engineering,Chongqing University,Chongqing 400030 China) |
| |
| Abstract: | A novel short-time spectrum estimation algorithm of speech enhancement based on distortion control is presented to overcome the speech distortion and low speech intelligibility in conventional method.An objective measurement parameter of speech distortion is introduced in the algorithm and a constraint condition is generated according to the parameter.Then,a modified log spectral amplitude estimation algorithm is proposed based on the speech absence probability and masking properties of the human auditory s... |
| |
| Keywords: | speech enhancement noise reduction speech intelligibility distortion control |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
