AI Resources

Open Source Models and Papers

Symbolic models

Folk-RNN, MusicVAE

https://ar5iv.labs.arxiv.org/html/2410.20515v1#bib.bib5#:~:text=Folk,large%20collection%20of%20folk%20tunes

Music Transformer

https://magenta.tensorflow.org/music-transformer#:~:text=Generating%20long%20pieces%20of%20music,performances%20generated%20by%20the%20model

MuseGAN

github.com

DeepBach

samgoree.github.io

Coconet

dash.harvard.edu

Museformer

github.com

PopMAG

slideshare.net

SongMASS

github.com

MuseCoco

arxiv.org

GETMusic

arxiv.org

MusicBERT

github.com

 

Audio generation

WaveNet

arxiv.org

SampleRNN

sander.ai

GANSynth

magenta.tensorflow.org

DDSP

magenta.tensorflow.org

DiffWave

arxiv.org

AudioGen

arxiv.org

MusicGen (long list of more links)

github.com.

 

Transcription

Onsets and Frames

magenta.tensorflow.org

MT3

openreview.net.

 

Tagging/Classification

musicnn

upf.edu

PANNs

arxiv.org

Tagging Transformer

mac.kaist.ac.kr

CLAP

medium.com

 

Source Separation

Spleeter

theoj.org

Open-Unmix

sigsep.github.io

Demucs

github.com.

 

Singing Synthesis

HiFiSinger

arxiv.org

DiffSinger

arxiv.org.