https://github.com/Ludwiggle/GRUMIDI

Recurrent Neural Network for generative MIDI music
https://github.com/Ludwiggle/GRUMIDI

algorave electronic-music gated-recurrent-units generative-art generative-music machine-learning mathematica midi midi-sequencer music recurrent-neural-networks wolfram-language wolfram-mathematica wolframlanguage wolframscript

Last synced: about 1 month ago
JSON representation

Recurrent Neural Network for generative MIDI music

• Host: GitHub
• URL: https://github.com/Ludwiggle/GRUMIDI
• Owner: Ludwiggle
• Created: 2018-08-08T14:30:02.000Z (almost 7 years ago)
• Default Branch: master
• Last Pushed: 2018-08-21T11:14:25.000Z (over 6 years ago)
• Last Synced: 2024-11-03T09:31:34.089Z (6 months ago)
• Topics: algorave, electronic-music, gated-recurrent-units, generative-art, generative-music, machine-learning, mathematica, midi, midi-sequencer, music, recurrent-neural-networks, wolfram-language, wolfram-mathematica, wolframlanguage, wolframscript
• Language: Mathematica
• Homepage:
• Size: 8.79 KB
• Stars: 4
• Watchers: 2
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# GRUMIDI

A GRU1-based RNN2 for rhythmic pattern generation.

The RNN model is a

[char-rnn](http://karpathy.github.io/2015/05/21/rnn-effectiveness/)

that gets trained on an input MIDI file encoded as a sequence of

[unit vectors](https://en.wikipedia.org/wiki/Unit_vector)

## Prerequisite

[WolframKernel](https://www.wolfram.com/cdf-player) 

[Wolframscript](https://www.wolfram.com/wolframscript)

Run `$ wolframscript -configure` and set the variable `WOLFRAMSCRIPT_KERNELPATH` to your local `WolframKernel` address

## Usage

1. Run `$ wolframscript -f encodeAndTrain.wl`

    Type the input filename5 `*.mid`

The trained net and decoding parameters are saved in `data/`.

2. Run `$ wolframscript -f generateAndDecode.wl`

    Generated `*.mid` is saved in `data/`.

## Discussion

In general, a MIDI file is not defined on a time-grid; MIDI events might be defined by machine-precision digits.

The first script will take care of time-quantization by fitting every MIDI event on a time-grid the resolution of which is equal to the minimum distance between two consecutive events that are found in the input MIDI file.

The generated MIDI inherits this time-quantization.

The dimension of the [unit vectors](http://reference.wolfram.com/language/ref/UnitVector.html) is equal to the number of different "notes" found in the input MIDI, e.g. the chromatic scale would be encoded with 12-dimensional unit vectors. Polyphony is encoded by vector addition of simultaneous events.

Similarly to [LSTMetallica](https://github.com/keunwoochoi/LSTMetallica), the encoded input MIDI is riffled with "BAR" every 16 unit vectors for *segmentation of measures*. These "BAR" markers are deleted once the nerual net output is decoded to MIDI format.

--------------------------------

1Gated Recurrent Unit

2Recurrent Neural Network

3Musical Instrument Digital Interface

4Full address or local address.