Recursively updated eigenfilterbank for speech enhancement

Authors:
  • CRC.A. Rodbro
    C.A. Rodbro
  • SJS.H. Jensen
    S.H. Jensen

Abstract
In this paper a novel signal subspace method for speech enhance- ment is proposed. The algorithm is derived from the filterbank interpretation of the truncated (quotient) singular value decompo- sition (T(Q)SVD) algorithm. We derive a recursive version of this algorithm which results in a recursively updated eigenfilterbank. The proposed method benefits from a low system delay and a low amount of musical noise in the enhanced speech signal.

This publication has 7 references indexed in Scilit: