About the framework
Welcome to the (SPARE) SParse Audio REstoration project website. You will find below some details about the modeling and algorithmic framework proposed to address audio restoration problems as denoising or declipping.
This work is developed by actual and former members of PANAMA (a joint project-team between INRIA and CNRS).
It is based on the expression of audio reconstruction scenarii as linear inverse problems and the use of analysis, synthesis plain or social sparse time-frequency priors.
Warning: use Firefox or Google Chrome browsers to play the sounds.
You can listen to denoising audio examples.
You can listen to declipping audio examples.
The examples shown in these two pages are from the RWC Database*.
*M. Goto, H. Hashiguchi, T. Nishimura, and R. Oka: RWC Music Database: Popular, Classical, and Jazz Music Databases In Proceedings of the 3rd International Conference on Music Information Retrieval (ISMIR 2002), 2002.
- C. Gaultier, N. Bertin, S. Kitić and R. Gribonval. A modeling and algorithmic framework for (non)social (co)sparse audio restoration. Submitted to a journal, 2017. [pre-print]
- C. Gaultier, S. Kitić, N. Bertin, R. Gribonval. AUDASCITY: AUdio Denoising by Adaptive Social CosparsITY. 25th European Signal Processing Conference (EUSIPCO), 2017.
- S. Kitić, N. Bertin and R. Gribonval. Sparsity and cosparsity for audio declipping: a flexible non-convex approach. In Latent Variable Analysis and Signal Separation, Liberec, 2015.
Visit the team website: PANAMA
Related project: spade
Do not hesitate to contact us through the form below for any inquiry (details about the framework, code, …)