AUDASCITY: AUdio Denoising by Adaptive Social CosparsITY - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2017

AUDASCITY: AUdio Denoising by Adaptive Social CosparsITY


This work aims at introducing a new algorithm, AUDASCITY, and comparing its performance to the time-frequency block thresholding algorithm for the ill-posed problem of audio denoising. We propose a heuristics which combines time-frequency structure, cosparsity, and an adaptive scheme to denoise audio signals corrupted with white noise. We report that AUDASCITY outperforms state-of-the-art for each numerical comparison. While there is still room for some perceptual improvements, AUDASCITY's usefulness is shown when used as a front-end for a classification task.
Fichier principal
Vignette du fichier
AudascityEUSIPCO_CR.pdf (319.36 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-01540945 , version 1 (16-06-2017)


  • HAL Id : hal-01540945 , version 1


Clément Gaultier, Srđan Kitić, Nancy Bertin, Rémi Gribonval. AUDASCITY: AUdio Denoising by Adaptive Social CosparsITY. 25th European Signal Processing Conference (EUSIPCO), Aug 2017, Kos, Greece. ⟨hal-01540945⟩
508 View
552 Download


Gmail Mastodon Facebook X LinkedIn More