Joint-sparse modeling for audio inpainting

Ichrak Toumi 1 Valentin Emiya 1
1 QARMA - éQuipe d'AppRentissage de MArseille
LIS - Laboratoire d'Informatique et Systèmes
Abstract : We exploit the common sparse structure between similar audio frames in order to reconstruct missing samples in audio signals. While joint-sparse models and related algorithms have been widely studied, one important challenge is to locate such similar frames in a fast way and when some samples are missing. We propose and compare several similarity measures dedicated to this task. We then show how this leads to better reconstruction results than when processing the audio frames independently.
Complete list of metadatas

https://hal-amu.archives-ouvertes.fr/hal-01881338
Contributor : Valentin Emiya <>
Submitted on : Tuesday, September 25, 2018 - 4:50:53 PM
Last modification on : Wednesday, September 26, 2018 - 1:22:25 AM

Identifiers

  • HAL Id : hal-01881338, version 1

Collections

Citation

Ichrak Toumi, Valentin Emiya. Joint-sparse modeling for audio inpainting. iTWIST: international Traveling Workshop on Interactions between low-complexity data models and Sensing Techniques, Nov 2018, Marseille, France. ⟨hal-01881338⟩

Share

Metrics

Record views

48