This page propose a live demonstration of our audio in-painting algorithm. We do not store any of your files after the processing.
We present a method to compensate for long duration data gaps in audio signals, in particular music. To achieve this task, a similarity graph is constructed, based on a short-time Fourier analysis of reliable signal segments, e.g. the uncorrupted remainder of the music piece, and the temporal regions adjacent to the unreliable section of the signal. A suitable candidate segment is then selected through a convex optimization scheme and smoothly inserted into the gap.
The article is availlable on arXiv and the code can be downloaded here.