首页    期刊浏览 2024年12月04日 星期三
登录注册

文章基本信息

  • 标题:Spectral Subtraction for Enhancement of Speech Quality
  • 本地全文:下载
  • 作者:Gitu Geevarghese ; Milind S. Shah
  • 期刊名称:International Journal of Electronics, Communication and Soft Computing Science and Engineering
  • 印刷版ISSN:2277-9477
  • 出版年度:2015
  • 卷号:4
  • 期号:Special 1
  • 出版社:IJECSCSE
  • 摘要:This paper describes a method for enhancing speechcorrupted by additive white Gaussian noise. The method is basedon the spectral noise subtraction method. The basic methodsubtracts an estimate of the noise power specspeech power spectrum, setting negative differences to zero,recombining the new power spectrum with the original phase,and then reconstructing the time waveform with the help ofIFFT. While this method reduces the broadband noise, it alsointroduces an annoying "musical noise". A modified spectralsubtraction method is used which consists of subtracting anoverestimate of the noise power spectrum, and preventing theresultant spectral components from going below a presetminimum level (spectral floor). Listening tests were performed todetermine the quality and intelligibility of speech enhanced bythis method.
  • 关键词:Speech enhancement; spectral subtraction; over;subtraction; spectral floor parameter
国家哲学社会科学文献中心版权所有