【摘要】 语音增强技术是解决噪声污染的一项强有力的预处理技术。谱减法通过处理后的语音中会留下所谓的"音乐噪声",针对这个问题,提出了一种多带谱相减与感觉加权相结合的语音增强方法。对带噪语音进行多带谱相减,并根据人的听觉掩蔽特性,对多带谱相减后的信号进行感觉加权,从而进一步降低背景噪声。在语音失真和噪声抑制之间取得良好的折中,减少语音的听觉失真,有效地抑制"音乐噪声",提高语音的清晰度。更多还原

【Abstract】 The pronunciation strengthened technology is a powerful treatment technology for solving the noise pollution.A speech enhancement method based on multi-band spectral subtraction plus the perceptual weighting algorithm is put forward.Firstly,this method performs multi-band spectral subtraction,and then based on the acoustical masking characteristics of human auditory system a processing of weight to feeling is brought forward to bear upon the subtracted signals.Test results show that a better compromise between reducing speech distortion and reinforcing noise suppression is made,speech distortion are decreased apparently,musical noise is suppressed and speech articulation is improved.更多还原