Skip to content

Simplify Installation Process #21

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
Ma5onic wants to merge 3 commits into
base: main
Choose a base branch
from
Open

Simplify Installation Process #21

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
c39e83c
Update README.md
Ma5onic Apr 13, 2023
39291cb
Further Simplify Install Process
Ma5onic Apr 13, 2023
78c6d7e
Update README.md
Ma5onic Jun 27, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
25 changes: 9 additions & 16 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
# Music Separation Enhancement With Generative Modeling

[![Open in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/interactiveaudiolab/MSG/blob/main/MSG.ipynb)`
[![Open in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/interactiveaudiolab/MSG/blob/main/MSG.ipynb)

This is the official implementation of the Make it Sound Good (MSG) model from our 2022 ISMIR paper "Music Separation Enhancement with Generative Modeling" [\[paper\]](https://arxiv.org/pdf/2208.12387.pdf)[ \[website\]](https://interactiveaudiolab.github.io/project/msg.html)

Expand All @@ -14,23 +14,16 @@ We introduce Make it Sound Good (MSG), a post-processor that enhances the output
- [Citation](#Citation)

## Setup
1. We train our model using salient source samples from the training data. To get the salient source samples, our training loop uses [nussl's](https://github.com/nussl/nussl/tree/salient_mixsrc2/nussl) SalientExcerptMixSourceFolder class from the salient_mixsrc2 branch. The specific branch of the repo can be downloaded using the steps below:
1. Download our repo from github.
```bash
git clone https://github.com/interactiveaudiolab/MSG.git
```
$ git clone https://github.com/nussl/nussl.git
$ cd nussl
$ git checkout salient_mixsrc2
$ pip install -e .
2. Change to the MSG repo folder and install the requirements.
```bash
cd MSG
pip install -r requirements.txt
```
2. Download our repo from github.
```
$ git clone https://github.com/interactiveaudiolab/MSG.git
```
3. Change to the MSG repo folder and download the requirements.txt.
```
$ cd MSG
$ pip install -r requirements.txt
```
4. If you would like to use our [pretrained checkpoints on huggingface download](https://huggingface.co/boazcogan/MSG_pretrained_checkpoints/tree/main) the model.
3. If you would like to use our [pretrained checkpoints on huggingface download](https://huggingface.co/boazcogan/MSG_pretrained_checkpoints/tree/main) the model.

## Training

Expand Down
1 change: 1 addition & 0 deletions requirements.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,3 +12,4 @@ SoundFile
librosa==0.8.1
torch
wandb
git+https://github.com/nussl/nussl.git@salient_mixsrc2