benzaiten

Description

These codes are for 2nd benzaiten contest.

Lakh MIDI Dataset(LMD-matched)
- This dataset is distributed with a CC BY 4.0.
- Colin Raffel. "Learning-Based Methods for Comparing Sequences, with Applications to Audio-to-MIDI Alignment and Matching". PhD Thesis, 2016.
MAESTRO Dataset
- This dataset is distributed with a CC BY-NC-SA 4.0.
- Curtis Hawthorne, Andriy Stasyuk, Adam Roberts, Ian Simon, Cheng-Zhi Anna Huang, Sander Dieleman, Erich Elsen, Jesse Engel, and Douglas Eck. "Enabling Factorized Piano Music Modeling and Generation with the MAESTRO Dataset." In International Conference on Learning Representations, 2019.
Infinite Bach
Weimar Jazz Database
- This dataset is distributed with a the Open Data Commons Open DataBase License (ODbL).
- Pfleiderer, Martin and Frieler, Klaus and Abeser, Jakob and Zaddach, Wolf-Georg and Burkhart, Benjamin. "Inside the Jazzomat - New Perspectives for Jazz Research" Schott Campus, 2017.
Charlie Parker's Omnibook MusicXML data
- This dataset is distributed with a CC BY-NC-SA 2.0.
- Ken Deguernel, Emmanuel Vincent, and Gerard Assayag. "Using Multidimensional Sequences for Improvisation in the OMax Paradigm", in Proceedings of the 13th Sound and Music Computing Conference, 2016.
OpenEWLD
- The compressed MusicXML dataset is distributed with a CC0.

bash bin/train.sh

bash bin/generate.sh

bash bin/generate_multiple.sh

These codes are licensed under MIT License without specific copyright holder.

Some codes are licensed under MIT License.