This project takes a set of music files and automatically orders them so that they are arranged to approximate how a professional might sequence them. Two approaches are demonstrated here. The first is the direct transformer-based approach presented in Automatic Album Sequencing by Vincent Herrmann, Dylan R. Ashley, and Juergen Schmidhuber. The second is the contrastive template-based approach presented in On the Distillation of Stories for Transferring Narrative Arcs in Collections of Independent Media by Dylan R. Ashley, Vincent Herrmann, Zachary Friggstad, and Juergen Schmidhuber.
Note that this project is forked from https://github.com/dylanashley/story-distiller and contains a lot of material directly copied from there.
This project is implemented in Python and uses models learned with PyTorch.
Before installation, first, ensure you have a recent version of Python (>=3.10) and pip, then clone the repository:
git clone git@github.com:dylanashley/automatic-album-sequencing.gitAfterwards, install the Python dependencies using pip:
pip install -r requirements.txtTo use the full suite of file types supported by the tool, it is also necessary to separately install the ffmpeg tool by following the instructions for your operating system provided on their website.
You can now run the web app by directly executing the app.py file, run the tool by directly executing the __main__.py file, or—if you're using Linux or macOS—you can use the makefile to compile the tool and then install it as an executable Python zip archive:
make all
sudo make installTo run the web app, simply execute the app.py file:
./app.pyTo run the program, execute it while passing the audio files as command-line arguments:
sdistil files [files ...] >> playlist.txtlibrosa is used to process the audio files, so most common audio file types are supported.
If you want to try out a different template, pass the -t argument to the program with the template file as an argument. Several learned templates are included in the templates directory:
