Our paper on Compressive Spectral Clustering has been accepted to ICML.

Read it on arXiv:1602.02018 or hal:hal-01320214

Download the code at http://cscbox.gforge.inria.fr

FaµST (pronounce FAUST) yields computationnaly efficient approximations to large dense matrices that can speed up iterative solvers for large-scale linear inverse problems.

See details on the methodology behind FaµST and some other applications, in our paper Le Magoarou L. and Gribonval R., “Flexible multi-layer sparse approximations of matrices and applications”, Journal of Selected Topics in Signal Processing, 2016.

Download the code at  http://faust.gforge.inria.fr/ and … stay tuned for the upcoming C++ FaµST library.

The French-speaking journal “Traitement du Signal” publishes this trimester a special issue on the theme of Multimodality and New Interactions in Music Signal Processing (“Traitement Des Signaux Musicaux Multimodalité Et Nouvelles Interactions”), coordinated by Nancy BERTIN, Frédéric BIMBOT, Jules ESPIAU DE LAMAËSTRE and Anaïk OLIVERO, members or former members of PANAMA team.

All articles published in this issue include a one-page abstract in English.

The PANAMA team @ Inria, Rennes, France is seeking highly qualified post-doctoral candidates  to contribute to the development of a general theoretic and algorithmic framework for compressive statistical learning, by leveraging concepts from compressive sensing and graph signal processing.

Candidates should hold a Ph.D. in applied mathematics, statistics, theoretical computer science, or mathematical signal processing.

Details and application form here.

Multi-channel BSS Locate is a Matlab toolbox to estimate direction of arrival (expressed both in azimuth and elevation) of multiple sources in a multi-channel audio signal recorded by an array of microphones. This toolbox implements the previous 8 angular spectrum methods available in BSS Locate toolbox.

Matlab toolbox:

http://bass-db.gforge.inria.fr/bss_locate/#mbss_locate

Online version (no Matlab required):

https://allgo.inria.fr/app/multichannelbsslocate

More informations on this page.

Scientific computing on large-scale data for signal processing and machine learning.

Details of the research engineer position at  http://team.inria.fr/panama/job-offers/

http://arxiv.org/abs/1511.05118

http://bit.ly/1NTMBBW

The PANAMA team (Parsimony and New Algorithms for Audio & Signal Modeling) at Inria, Rennes, France is seeking highly qualified candidates for postdoctoral positions funded by the ERC (European Research Council) in the context of the project PLEASE (Projection, Learning and Sparsity for Efficient data processing). Appointments are intended to start at the beginning of 2016.

For more information, visit the announcement page: http://team.inria.fr/panama/please-people/