Music-Code

Music-Code is a python library for creating music. You can create musical notes, chords, progressions, melodies, bass lines, drum beats, sound design and full songs! It has most of the core features of a digital audio workstation (DAW). In addition to creating music, you can also generate HD waveform visualizations for images and video. Music-Code is integrated with MySQL and provides an analytics dashboard and Rest API to monitor system usage and data trends.

You can use Music-Code to automatically produce large audio datasets for use in deep learning applications. You can scan through different spaces of musical/tonal combinations, exporting each variation as a labeled WAV file. The long term purpose of Music-Code is to create vast musical datasets and build music AI.

Setup

Download Program Files

First, download the Music-Code program files here: Music-Code Program Files. Unzip and save the program files in a memorable location. Then copy the file path (you will need this in a moment).

Create Anaconda Environment

Anaconda is the ideal way to run Music-Code.

1. Install Anaconda on your machine
2. Create new conda environment named music_code
3. Activate music_code environment
4. Install the following requirements:

Installation

pip install music-code

Requirements

numpy==1.18.4 pandas==1.0.3 matplotlib==3.2.1 seaborn==0.10.1 soundfile==0.10.3.post1 scipy==1.4.1 sklearn==0.0 datetime==4.3 PyAudio==0.2.11 wave==0.0.2 mysql-connector-python==8.0.20 psutil==5.7.0

Set Program Files Path

1. Open your Anaconda3 folder, and locate the music_code installation. For my machine, the file is located here: C:\Users\wesle\Anaconda3\envs\music_code\Lib\site-packages\music_code\music_code.py
2. Open music_code.py, go to line 42, in the MusicCode init function, find the program_files_location attribute. Copy and paste the file path to your program files folder, which you can download here: Music-Code Program Files. Once this path is set, the Music-Code file system is good to go. The program files folder contains the Music-Code sample library and datasets. All the WAV files and images you create are stored in this folder.

MySQL (optional)

Connect Music-Code to a MySQL database to archive all system data and have access to an analytics dashboard. To create the Music-Code MySQL database, see the create_database.sql file.

Test

Run the tests.py file inside your conda environment to ensure the all systems are working properly.

Jupyter Notebook Tutorials

Check out the jupyter notebook tutorials. These demonstrate all the capabilities of the Music-Code library.

YouTube Tutorial

Soon I will post a video about geting started with Music-Code!

Contact

For any other questions, email wesleylaurencetech@gmail.com