festinuz/denoising-historical-data
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
9 Commits 1 Branch 0 Tags 162 KiB
Python 68%
Jupyter Notebook 31.6%
Shell 0.4%
535a0d2b58
Go to file
eloimoliner 535a0d2b58
Update README.md 
updatin readmer
2021-08-31 15:48:32 +03:00
conf removing test 2021-08-31 11:16:16 +03:00
experiments/trained_model uploading even more shit 2021-08-30 18:43:40 +03:00
dataset_loader.py dding neviroment 2021-08-31 15:36:06 +03:00
environment.yml dding neviroment 2021-08-31 15:36:06 +03:00
inference.py dding neviroment 2021-08-31 15:36:06 +03:00
inference.sh preparing files 2021-08-31 15:41:56 +03:00
README.md Update README.md 2021-08-31 15:48:32 +03:00
requirements.txt dding neviroment 2021-08-31 15:36:06 +03:00
train.py dding neviroment 2021-08-31 15:36:06 +03:00
train.sh preparing files 2021-08-31 15:41:56 +03:00
trainer.py dding neviroment 2021-08-31 15:36:06 +03:00
unet.py startin again 2021-08-30 18:30:51 +03:00

README.md

A two-stage U-Net for high-fidelity denoising of historical recordings

Official repository of the paper:

citation here

Abstract

abstract here

Schema represention

Requirements

conda env update -f environment.yml
conda activate historical_denoiser

Denosing Recordings

Training