摘要
提出了一种谱减法的改进形式,算法打破了噪声和语音是相互独立且噪声是零均值的高斯分布假设。实验表明这种改进型谱减算法有效提高了增强效果,更好地抑制了音乐噪声。
An improved spectral subtraction algorithm is presented. This algorithm breaks the assumption that noise and speech are independent from each other and noise is Gaussian distribution with mean O. Experimental results show that the proposed algorithm improves the speech quality and reduces the musical noise effectively.
出处
《电声技术》
2008年第2期46-48,共3页
Audio Engineering
关键词
谱减法
语音增强
音乐噪声
spectral subtraction algorithm
speech enhancement
musical noise
