Skip to main content

Learnable low rank sparse models for speech denoising

Publication ,  Journal Article
Sprechmann, P; Bronstein, A; Bronstein, M; Sapiro, G
Published in: ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
October 18, 2013

In this paper we present a framework for real time enhancement of speech signals. Our method leverages a new process-centric approach for sparse and parsimonious models, where the representation pursuit is obtained applying a deterministic function or process rather than solving an optimization problem. We first propose a rank-regularized robust version of non-negative matrix factorization (NMF) for modeling time-frequency representations of speech signals in which the spectral frames are decomposed as sparse linear combinations of atoms of a low-rank dictionary. Then, a parametric family of pursuit processes is derived from the iteration of the proximal descent method for solving this model. We present several experiments showing successful results and the potential of the proposed framework. Incorporating discriminative learning makes the proposed method significantly outperform exact NMF algorithms, with fixed latency and at a fraction of it's computational complexity. © 2013 IEEE.

Duke Scholars

Published In

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

DOI

ISSN

1520-6149

Publication Date

October 18, 2013

Start / End Page

136 / 140
 

Citation

APA
Chicago
ICMJE
MLA
NLM
Sprechmann, P., Bronstein, A., Bronstein, M., & Sapiro, G. (2013). Learnable low rank sparse models for speech denoising. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, 136–140. https://doi.org/10.1109/ICASSP.2013.6637624
Sprechmann, P., A. Bronstein, M. Bronstein, and G. Sapiro. “Learnable low rank sparse models for speech denoising.” ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, October 18, 2013, 136–40. https://doi.org/10.1109/ICASSP.2013.6637624.
Sprechmann P, Bronstein A, Bronstein M, Sapiro G. Learnable low rank sparse models for speech denoising. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. 2013 Oct 18;136–40.
Sprechmann, P., et al. “Learnable low rank sparse models for speech denoising.” ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, Oct. 2013, pp. 136–40. Scopus, doi:10.1109/ICASSP.2013.6637624.
Sprechmann P, Bronstein A, Bronstein M, Sapiro G. Learnable low rank sparse models for speech denoising. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. 2013 Oct 18;136–140.

Published In

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

DOI

ISSN

1520-6149

Publication Date

October 18, 2013

Start / End Page

136 / 140