search

andabi / music-source-separation

Deep neural networks for separating singing voice from music written in TensorFlow
795 stars 150 forks source link

Instructions for use within Ubuntu 18.04. #25

Open 0xBEEEF opened 5 years ago

0xBEEEF commented 5 years ago

I would like to try this project, but unfortunately there is no "real" guide on how to set it up. What packages do I need in Ubuntu 18.04? I'm not familiar with TensorFlow or any other machine learning program yet. Therefore I find the entry hurdle relatively high. Could one perhaps provide a slightly better setup guide? At the moment there is more information about the algorithm than about the actual setup.

Many thanks for your help in advance!

0xBEEEF commented 5 years ago

For information only. I got it meanwhile to install everything under Ubuntu.

I only had problems finding the right Python version. The use of the standard PIP as described in the ReadMe unfortunately didn't work. Instead I had to use pip3, which was actually intended for Python 3. Then there was one entry missing in the Requirements.txt file. I added this one, and then it worked.

I leave the point open again, as a reminder for the developers to adjust the ReadMe, and to check the file Requirements.txt.