eduzhai > Applied Sciences > Engineering >

Efficient Independent Vector Extraction of Dominant Target Speech

  • king
  • (0) Download
  • 20210507
  • Save

... pages left unread,continue reading

Document pages: 5 pages

Abstract: The complete decomposition performed by blind source separation iscomputationally demanding and superfluous when only the speech of one specifictarget speaker is desired. In this paper, we propose a computationallyefficient blind speech extraction method based on a proper modification of thecommonly utilized independent vector analysis algorithm, under the mildassumption that the average power of signal of interest outweighs interferingspeech sources. Considering that the minimum distortion principle cannot beimplemented since the full demixing matrix is not available, we also design aone-unit scaling operation to solve the scaling ambiguity. Simulations validatethe efficacy of the proposed method in extracting the dominant speech.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×