This repository contains the Score Follower application developed as part of a master's project, published on arXiv at https://arxiv.org/abs/2502.10426. This product serves as a first application which uses Gaussian Processes to perform score following.
- The score follower is designed for evaluation using the adapted score renderer app available here: https://github.com/josephinecowley/Score-follower.
- A pre-recorded .wav file if running in Pre-recorded Mode.
- Score in MIDI format.
- Score in MusicXML format.
- Dependencies listed in
requirements.txt
- Download the adapted Flippy Qualitative Testbench in the Prerequisites section.
- Update the parameters and GP hyperparameters in the file
args.py. - Follow instructions in https://github.com/flippy-fyp/flippy-qualitative-testbench/blob/main/README.md to use the score renderer as an evaluation tool of this score follower.
-
python scorefollow.py
Adjust the command with the appropriate flags or input parameters as needed for your specific setup.
Contributions to the Score Follower project are welcome. Please follow the steps below to contribute:
- Fork the repository.
- Create a new branch for your feature (
git checkout -b feature/your_feature_name). - Commit your changes (
git commit -am 'Add some feature'). - Push to the branch (
git push origin feature/your_feature_name). - Create a new Pull Request.
The architecture of lib/ is based on Flippy, though the implementation of each significant component is original.