Skip to content

Code and Datasets for the "Deep Learning for Music Classification using Keras" Workshop

Notifications You must be signed in to change notification settings

slychief/mlprague2018_tutorial

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

46 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Deep Learning for Music Classification using Keras

(c) 2018 by

Alexander Schindler, AIT Austrian Institute of Technology
Website: http://ifs.tuwien.ac.at/~schindler
Twitter: https://twitter.com/Slychief

Thomas Lidy, Musimap
Website: https://www.linkedin.com/in/thomaslidy/
Twitter: https://twitter.com/LidyTom

also visit: https://www.meetup.com/Vienna-Deep-Learning-Meetup

Abstract

Deep Learning has re-entered the field of Machine Learning with a big bang. Tasks that were difficult and cumbersome to solve are now simple to implement and achieve amazingly high accuracies. Deep Learning has shown substantial impact recently also in the domain of audio recognition and music classification. This tutorial will give a general introduction to Neural Networks, Convolutional Neural Networks, plus important concepts such as (batch-)normalization, training epochs, activation, loss, optimization, pooling, dropout, model fine-tuning, etc. and a tutorial on how to apply all of this on tasks of audio and music recognition using the convenient Python Keras framework on top of Tensorflow.

Tutorial Outline

Part 1 - Deep Learning Basics (Slides)

Part 2 - Convolutional Neural Networks (Slides)

  • Difference CNN – RNN
  • How CNNs work (Layers, Filters, Pooling)
  • Application Domains and how to use in Music
  • Coding Examples (Jupyter Notebook)

Part 3 - Instrumental, Genre and Mood Analysis (Slides)

  • Large-scale Music AI at Musimap
  • Instrumental vs. Vocal Detection
  • Genre Recognition
  • Mood Detection
  • Coding Examples (Jupyter Notebook)

Part 4 - Advanced Deep Learning

Tutorial Requirements

For the tutorials, we use iPython / Jupyter notebook, which allows to program and execute Python code interactively in the browser.

Viewing Only

If you do not want to install anything, you can simply view the tutorials' content in your browser, by clicking on the tutorial's filenames listed above in the GIT file listing.

The tutorials will open in your browser for viewing.

Interactive Coding

If you want to follow the tutorials by actually executing the code on your computer, please install first the pre-requisites as described below.

After that, to run the tutorials go into the mlprague2018_tutorial folder and start from the command line:

jupyter notebook

Installation of Pre-requisites

Install Python 3.x

Note: On most Mac and Linux systems Python is already pre-installed. Check with python --version on the command line whether you have Python 3.x installed.

Otherwise install Python 3.5 from https://www.python.org/downloads/release/python-350/

Install Python libraries:

Mac, Linux or Windows

(on Windows leave out sudo)

Important note: If you have Python 2.x and 3.x installed in parallel, replace pip by pip3 in the following commands:

sudo pip install --upgrade jupyter

Try if you can open

jupyter notebook

on the command line.

Then download or clone the Tutorials from this GIT repository:

git clone https://github.com/slychief/mlprague2018_tutorial.git

or download https://github.com/slychief/mlprague2018_tutorial/archive/master.zip
unzip it and rename the folder to mlprague2018_tutorial.

Install the remaining Python libraries needed:

Either by:

sudo pip install Keras>=2.1.1 tensorflow scikit-learn>=0.18 pandas librosa spotipy matplotlib

or, if you downloaded or cloned this repository, by:

cd mlprague2018_tutorial
sudo pip install -r requirements.txt

Optional for GPU computation

If you want to train your neural networks on your GPU (which is faster, but not necessarily needed for this tutorial), you have to install the specific GPU version of Tensorflow:

sudo pip install tensorflow-gpu

and also install the following:

Install Audio Decoder

In order to decode MP3 files (used in the MagnaTagAtune data set) you will need to install FFMpeg on your system.

Download Prepared Datasets

Please download the following data sets for this tutorial:

GTZAN Music Speech Classification

https://owncloud.tuwien.ac.at/index.php/s/JiBXUPZK9LImTHB (145MB)

MagnaTagAtune

https://owncloud.tuwien.ac.at/index.php/s/hivOGXKoUQtacbo (332MB)

These are prepared versions from the original datasets described below.

Credits

The following helper Python libraries are used in these tutorials:

  • audiofile_read.py and audio_spectrogram.py: by Thomas Lidy and Alexander Schindler, taken from the RP_extract git repository
  • wavio.py: by Warren Weckesser

The data sets we use in the tutorials are from the following sources:

(don't download them from there but use the prepared datasets from the two owncloud links above)

About

Code and Datasets for the "Deep Learning for Music Classification using Keras" Workshop

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published