🎶 LUC on SoundCloud 🎶
NOTE: Do not forget to unmute the player below to hear the music
LUC-Main-Sample.mp4
Model trained on 70951 Pitches-Drums pairs from clean_midi/LAKH MIDI Datasets
Clean MIDI Transformer Model Raw Training Stats
Epoch: 1 Loss: 0.02231 LR: 0.00012121694: 100%|██████████| 132924/132924 [2:49:01<00:00, 13.11it/s]
Loss val: 0.01247 Acc: 0.9957: 23%|██▎ | 922/3988 [00:31<01:43, 29.57it/s]
The Lakh MIDI Dataset is distributed with a CC-BY 4.0 license; if you use this data in any capacity, please reference this page and my thesis:
Colin Raffel. "Learning-Based Methods for Comparing Sequences, with Applications to Audio-to-MIDI Alignment and Matching". PhD Thesis, 2016.
Of course, I did not transcribe any of the MIDI files in the Lakh MIDI Dataset. While MIDI files have a built-in mechanism for attribution (the Copyright meta-event), it is not used consistently, so attributing each of the MIDI files in the dataset to a particular author is not feasible.
@inproceedings{lev2021larsulrichchallenge,
title = {Lars Ulrich Challenge},
author = {Aleksandr Lev},
booktitle = {GitHub},
year = {2021},
}