Inconsistent Audio Declipping Performance Enhancement Based on Audio Inpainting
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26220%2F21%3APU141180" target="_blank" >RIV/00216305:26220/21:PU141180 - isvavai.cz</a>
Result on the web
<a href="https://www.fekt.vut.cz/conf/EEICT/archiv/sborniky/EEICT_2021_sbornik_1.pdf" target="_blank" >https://www.fekt.vut.cz/conf/EEICT/archiv/sborniky/EEICT_2021_sbornik_1.pdf</a>
DOI - Digital Object Identifier
—
Alternative languages
Result language
angličtina
Original language name
Inconsistent Audio Declipping Performance Enhancement Based on Audio Inpainting
Original language description
Some of the state-of-the-art audio declipping methods are not consistent with the observed signal, meaning that they do not keep the a priori undegraded (reliable) samples. These samples can be directly substituted in the reconstructed signals, but this may create audible artifacts due to the sharp transitions between the declipped and the substituted parts. We propose two methods based on audio inpainting, which deal with these transitions. As a result, we observe a significant improvement of the PEAQ ODG values, but only for some of the declipping algorithms considered.
Czech name
—
Czech description
—
Classification
Type
O - Miscellaneous
CEP classification
—
OECD FORD branch
20203 - Telecommunications
Result continuities
Project
<a href="/en/project/GA20-29009S" target="_blank" >GA20-29009S: Perceptually motivated restoration of highly degraded audio signals</a><br>
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)
Others
Publication year
2021
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů