Audio_declipping

This repository wraps all functions and scripts for running results in 'Final_report.pdf'.

In order to run the evaluation of the models, run the script Models_evaluation.ipynb. To be able to run the script, you have to adapt the paths to the data. The data can be downloaded from https://www.openslr.org/12, and choosing the 'dev-clean.tar.gz' file.

You can run the script 'run_example' to load a single track, perform de-clipping, and plot and hear the results of the DEMCS model and the A-SPADE algorithm.

The models themselves are stored in this repository, under the "Models" folder.