Symbolic Music Generation

This repository contains the implementation for Homework 3 of the CommE5070 Deep Learning for Music Analysis and Generation course, Fall 2024, at National Taiwan University. For a detailed report, please refer to this slides.

Setup

To set up the virtual environment and install the required packages, use the following commands:

virtualenv --python=python3.10 deepmir_hw3
source deepmir_hw3/bin/activate
pip install -r requirements.txt

Data and Checkpoint Download

Dataset

To download the dataset, run the following script:

bash scripts/download_data.sh

Checkpoint

To download the pre-trained model checkpoints, use the command:

bash scripts/download_ckpt.sh

Sound Font

To download the sound font, use the command:

bash scripts/download_soundfont.sh

Training

To train the model, run the command:

python train.py

Inference

Task 1

To generate the symbolic music, run the command:

python task1.py \
    --ckpt_path checkpoints/<checkpoint name> \
    --output_folder <Folder path for saving the generated file>

Task 2

To continue the symbolic music, run the command:

python task2.py \
    --ckpt_path checkpoints/<checkpoint name> \
    --prompt_song_path <Folder path for prompt song> \
    --output_folder <Folder path for saving the generated file>

Evaluation

To evaluate the generated results, run the command:

python eval.py \
    --eval_folder <Results folder to evaluation> \
    --output_path <Path for output score.csv>

Converting MIDI Files to WAV Format

To convert .mid file to .wav file, run the command:

python convert_mid_to_wav.py \
    --input_folder <Path for input folder> \
    --output_folder <Path for output folder>

Environment

We implemented the code on an environment running Ubuntu 22.04.1, utilizing a 12th Generation Intel(R) Core(TM) i7-12700 CPU, along with a single NVIDIA GeForce RTX 4090 GPU equipped with 24 GB of dedicated memory.

Citation

If you use this code, please cite the following:

@misc{liao2024_symbolic_music_generation,
    title  = {Symbolic Music Generation},
    author = {Jia-Wei Liao},
    url    = {https://github.com/jwliao1209/Symbolic-Music-Generation},
    year   = {2024}
}