EI / SCOPUS / CSCD 收录

中文核心期刊

用于语音增强的高频信噪比度量

SNRH measurement for speech enhancement

  • 摘要: 一般使用信噪比的提高评估带噪语音的增强算法效果,本文提出一种新的信噪比计算方法——高频信噪比(SNRH),即用经过高通滤波的各信号计算信噪比。实验表明新方法比传统方法能更准确地评估语音增强算法的好坏和有效范围,是更合理的语音增强算法度量。

     

    Abstract: Improvement of SNR is generally used as the effect of speech enhancement. This paper suggests a modified SNR algorithm - SNRH (Signal-Noise-Ratio of Highpass) for speech enhancement. Experiments show the new method outperforms the traditional SNR as it can offer more information regarding the applied conditions of enhancement approaches and their relative efficiency.

     

/

返回文章
返回