eduzhai > Applied Sciences > Engineering >

Blind Mask to Improve Intelligibility of Non-Stationary Noisy Speech

  • king
  • (0) Download
  • 20210506
  • Save

... pages left unread,continue reading

Document pages: 5 pages

Abstract: This letter proposes a novel blind acoustic mask (BAM) designed to adaptivelydetect noise components and preserve target speech segments in time-domain. Arobust standard deviation estimator is applied to the non-stationary noisyspeech to identify noise masking elements. The main contribution of theproposed solution is the use of this noise statistics to derive an adaptiveinformation to define and select samples with lower noise proportion. Thus,preserving speech intelligibility. Additionally, no information of the targetspeech and noise signals statistics is previously required to this non-idealmask. The BAM and three competitive methods, Ideal Binary Mask (IBM), TargetBinary Mask (TBM), and Non-stationary Noise Estimation for Speech Enhancement(NNESE), are evaluated considering speech signals corrupted by threenon-stationary acoustic noises and six values of signal-to-noise ratio (SNR).Results demonstrate that the BAM technique achieves intelligibility gainscomparable to ideal masks while maintaining good speech quality.

Please select stars to rate!


0 comments Sign in to leave a comment.

    Data loading, please wait...