作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2006, Vol. 32 ›› Issue (24): 223-225. doi: 10.3969/j.issn.1000-3428.2006.24.081

• 多媒体技术及应用 • 上一篇    下一篇

语音增强算法评估的研究

孟 静,许 刚   

  1. (华北电力大学信息工程系,北京 102206)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2006-12-20 发布日期:2006-12-20

Research on Evaluation of Speech Enhancement Algorithms

MENG Jing,XU Gang   

  1. (Dep. of Information Engineering, North China Electric Power University, Beijing 102206)
  • Received:1900-01-01 Revised:1900-01-01 Online:2006-12-20 Published:2006-12-20

摘要: 研究了语音增强算法性能的评估方法。利用客观音质评价方便快捷的优点,针对SNR评估不能充分反映语音信号时变特征的不足,提出了采用IS、LLR、SegSNR 3种客观失真测度与SNR相结合的评估方法。实验表明,该方法可以有效地对语音增强算法进行整体和局部性能的评估。

关键词: 语音增强, 算法评估, 客观测度

Abstract: This paper explores the evaluation of speech enhancement algorithm performance. The approach it uses is to compute the distance measure of the original and degraded speech signal and then accomplishes the evaluation of speech quality according to the distance measure. It selects several speech enhancement algorithms and compares their performances through comparing the distance measure. In the end, it concludes an effective evaluation for speech enhancement.

Key words: Speech enhancement, Algorithm evaluation, Distance measurement