Skip to content

burliEnterprises/tensorflow-music-generator

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

17 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

tensorflow-music-generator


generates music (midi files) using a Tensorflow RNN

Read more about it on Medium: https://medium.com/@m_ko/deep-learning-with-tensorflow-part-3-music-and-text-generation-370cf37bb071

Hear a demo of the result here: https://www.youtube.com/watch?v=5wZIQ9XHNns&feature=youtu.be

Overview

Use TensorFlow to generate short sequences of music with a Restricted Boltzmann Machine. Original code comes from YouTube, see here: (https://youtu.be/ZE7qWXX05T0)

Dependencies

  • Tensorflow
  • pandas
  • numpy
  • msgpack-python
  • glob2
  • tqdm
  • python-midi

Use pip to install any missing dependencies (pip install --upgrade ... )

Dependencies on Windows with python3

    pip3 install pandas
    pip3 install msgpack-python
    pip3 install numpy
    pip3 install glob2
    pip3 install tqdm
    pip3 install py-midi

Basic Usage

To train the model and create music, simply clone this directory and run

python rbm_chords.py

The training data goes in the pop_music_midi folder. You have to use MIDI files. You can find some here. I have already added pop songs. Training will take 5-10 minutes on a modern laptop. The output will be a collection of midi files.

Credits

The credit for this code goes to dshieble and llSourcell

Have fun!