MLM_decoding

This is a working repository for our project on Music Language Model decoding.

In this repository, we make available the code to reproduce the experiments from some publications, in different branches:

Adrien Ycart, Andrew McLeod, Emmanouil Benetos and Kazuyoshi Yoshii. "Blending Acoustic and Language Model Predictions for Automatic Music Transcription", 20th International Society for Music Information Retrieval Conference (ISMIR), November 2019, Delft, Netherlands.
- Go to this branch: ISMIR2019

Name		Name	Last commit message	Last commit date
Latest commit History 466 Commits
blending_model/optim		blending_model/optim
hmm		hmm
mlm_training		mlm_training
paper/roadmap		paper/roadmap
.gitignore		.gitignore
README.md		README.md
beam.py		beam.py
beats_utils.py		beats_utils.py
comparison_plot_ismir.py		comparison_plot_ismir.py
copy_MAPS_wav_files.py		copy_MAPS_wav_files.py
dataMaps.py		dataMaps.py
datasetMaps.py		datasetMaps.py
decode.py		decode.py
eval_utils.py		eval_utils.py
evaluate.py		evaluate.py
evaluate_beats.py		evaluate_beats.py
evaluate_from_midi.py		evaluate_from_midi.py
evaluate_load.py		evaluate_load.py
evaluate_mlm.py		evaluate_mlm.py
explore_results.py		explore_results.py
listen_example.py		listen_example.py
plot_figures.py		plot_figures.py
plots_thesis.py		plots_thesis.py
sampling.py		sampling.py
scatter_plot_ismir.py		scatter_plot_ismir.py
state.py		state.py
tests_beat_steps.py		tests_beat_steps.py
train_branching_factor.py		train_branching_factor.py
train_mlm.py		train_mlm.py

Provide feedback