Skip to content

Code of the paper "Impact of time and note duration tokenizations on deep learning symbolic music modeling" (ISMIR 2023)

Notifications You must be signed in to change notification settings

Natooz/music-modeling-time-duration

Repository files navigation

Impact of time and note duration tokenizations on deep learning symbolic music modeling

Code of the paper Impact of time and note duration tokenizations on deep learning symbolic music modeling. (ISMIR 2023)

In this work, we analyze the current common tokenization methods and experiment with time and note duration representations. We compare the performance of these two impactful criteria on several tasks, including composer classification, emotion classification, music generation, and sequence representation.

Steps to reproduce

  1. pip install -r requirements to install requirements
  2. sh scripts/download_datasets.sh to download the POP909 and EMOPIA datasets;
  3. Download the GiantMIDI dataset and put it in data/
  4. python scripts/tokenize_datasets.py to tokenize data and learn BPE
  5. python exp_generation.py to train generative models and generate results
  6. python exp_pretrain.py to pretrain classification and contrastive models
  7. python exp_cla_finetune.py to train classification models and test them
  8. python exp_contrastive.py to train contrastive models and test them

About

Code of the paper "Impact of time and note duration tokenizations on deep learning symbolic music modeling" (ISMIR 2023)

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages