• v1.3: MIDITokenizerV2 and new MidiVisualizer
• v1.2 : Optimise the tokenizer and dataset. The dataset was filtered by MIDITokenizer.check_quality. Using the higher quality dataset to train the model, the performance of the model is significantly improved.

• online: huggingface

• online: colab

• download windows app

huggingface

projectlosangeles/Los-Angeles-MIDI-Dataset

• install pytorch(recommend pytorch>=2.0)
• install fluidsynth>=2.0.0
• pip install -r requirements.txt

python app.py

python train.py

@misc{skytnt2024midimodel,
  author = {SkyTNT},
  title = {Midi Model: Midi event transformer for symbolic music generation},
  year = {2024},
  howpublished = {\url{https://github.com/SkyTNT/midi-model}},
}