Sparse models and convex optimisation for convolutive blind source separation
Event details
| Date | 08.03.2011 |
| Hour | 14:00 |
| Speaker | Dr. Prasad Sudhakar, INRIA Rennes-Bretagne Atlantique, France |
| Location | |
| Category | Conferences - Seminars |
We jointly exploit the sparsity of the sources and mixing filters for blind estimation of sparse filters from stereo convolutive mixtures of several sources. First, we show why the sparsity of the filters can help solve the permutation problem in convolutive source separation, in the absence of scaling. Then, we propose a two-stage estimation framework, which is primarily based on the time-frequency domain cross-relation and an $ell^1$ minimisation formulation: a) a clustering step to group the time-frequency points where only one source is active, for each source; b) a convex optimisation step which estimates the filters. The resulting algorithms are assessed on audio source separation and filter estimation problems.
Practical information
- General public
- Free