Regularized non-negative matrix factorization with temporal dependencies for speech denoising

Kevin W. Wilson, Bhiksha Raj, Paris Smaragdis

Research output: Contribution to journalArticlepeer-review


We present a tecchnique for denoising speech using temporally regularized nonnegative matrix factorization (NMF). In previous work [1], we used a regularized NMF update to impose structure within each audio frame. In this paper, we add frame-to-frame regularization across time and show that this additional regularization can also improve our speech denoising results. We evaluate our algorithm on a range of nonstationary noise types and outperform a state-of-the-art Wiener filter implementation.

Original languageEnglish (US)
Pages (from-to)411-414
Number of pages4
JournalProceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH
StatePublished - 2008
Externally publishedYes


  • Source separation
  • Speech enhancement
  • Speech modeling
  • Speech processing

ASJC Scopus subject areas

  • Human-Computer Interaction
  • Signal Processing
  • Software
  • Sensory Systems


Dive into the research topics of 'Regularized non-negative matrix factorization with temporal dependencies for speech denoising'. Together they form a unique fingerprint.

Cite this