摘要
提出了一种有效的消除噪声且减小语音失真的语音增强方法。首先实现了语音信号服从Laplacian分布、噪声服从Gaussian分布假设下的MMSE增强算法。为了进一步提高语音增强效果,在增强语音谱幅度阈值的计算上将该方法与人的掩蔽特性相结合。通过语音增强方法性能客观评测表明,该语音增强方法更好地抑制了噪声,有效地减小语音失真。
An effective approach for attenuating acoustic noise and mitigating speech distortion is proposed.First,MMSE method is analysed when the clean speech is modeled by a Laplacian distribution and the noise is modeled by a Gaussian distribution. Then,human perceptual auditory masking threshold is incorporated into this approach when the threshold of spectral amplitude of enhanced speech is computed.The experiment result evaluated by objective more significant noise reduction and reduce the chances of speech distortion.
出处
《计算机工程与应用》
CSCD
北大核心
2010年第7期153-154,161,共3页
Computer Engineering and Applications
关键词
语音增强
听觉掩蔽阈值
最小均方误差(MMSE)
speech enhancement
masking properties
Minimum Mean-Square measure shows the proposed method can achieve a Error(MMSE)