Sparsity and low-rank amplitude based blind Source Separation

Fangchen Feng 1 Matthieu Kowalski 1
1 L2S - Laboratoire des signaux et systèmes
Abstract : This paper presents a new method for blind source separation problem in reverberant environments with more sources than microphones. Based on the sparsity property in the time-frequency domain and the low-rank assumption of the spectrogram of the source, the STRAUSS (SparsiTy and low-Rank AmplitUde based Source Separation) method is developed. Numerical evaluations show that the proposed method outperforms the existing multichannel NMF approaches, while it is exclusively based on amplitude information.
Keywords : Blind source separation sparsity low-rank multichannel non-negative matrix factorization
Document type :
Conference papers
Complete list of metadatas

Cited literature [25 references]  Display  Hide  Download
https://hal.archives-ouvertes.fr/hal-01547459
Contributor : Matthieu Kowalski <>
Submitted on : Monday, June 26, 2017 - 5:05:56 PM
Last modification on : Saturday, September 8, 2018 - 4:24:02 PM
Long-term archiving on : Wednesday, January 17, 2018 - 7:17:16 PM

File

FK_icassp17.pdf
Files produced by the author(s)

Identifiers

Collections

CNRS | L2S | CENTRALESUPELEC | SUP_LSS | UNIV-PARIS-SACLAY | SUP_SIGNAUX

Citation

Fangchen Feng, Matthieu Kowalski. Sparsity and low-rank amplitude based blind Source Separation. The 42nd IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2017), Mar 2017, New Orleans, United States. pp.571 - 575, ⟨10.1109/ICASSP.2017.7952220⟩. ⟨hal-01547459⟩

Export

BibTeX TEI DC DCterms EndNote

Share

Metrics

Record views

160

Files downloads

459

Contact

CCSD