The present tutorials are designed to implement the core notions seen in machine learning lessons. Most techniques can be applied to any type of data from which sets of features can be computed. The exercises here will introduce the basic mecanisms behind these technics and then, we will target specific applications to musical or audio data.

Get the baseline code for your language of choice :

Matlab zip file

Python zip file

Get the audio datasets from this link

Unzip files and place 00_Datasets along with other folders

Download the slides

• Introduction to artificial intelligence
• Properties of machine learning
• Nearest-neighbors

In this introduction, we will cover basic Music Information Retrieval (MIR) interactions, in which we process a dataset of sound files and try to observe the properties of their various temporal and spectral features. Hence, we will quickly review basic calculus required to perform further machine learning tasks. This tutorial is also intended to review basic Matlab coding and plotting operations.

Along the tutorials, we provide a reference code for each section. This code contains helper functions that will alleviate you from the burden of data import and other sideline implementations. You will find designated spaces in each file to develop your solutions. The code is in Python (notebooks impending) and relies heavily on the concept of code sections which allows you to evaluate only part of the code (to avoid running long import tasks multiple times and concentrate on the question at hand.

Get the baseline MATLAB code for all tutorials from this zip file

Get the baseline Python code for all tutorials from this zip file

Get the baseline Jupyter notebooks code for all tutorials from this zip file

Python installation

In order to get the baseline script to work, you need to have a working distribution of Python 3.5 as a minimum (we also recommend to update your version to Python 3.7). We will also be using the following libraries

• Matplotlib
• Numpy
• SciPy
• Scikit-Learn
• Music21
• Librosa
• PyTorch

We highly recommend that you install Pip or Anaconda that will manage the automatic installation of those Python libraries (along with their dependencies). If you are using Pip, you can use the following commands

pip install matplotlib
pip install numpy
pip install scipy
pip install scikit-learn
pip install music21
pip install librosa
pip install torch torchvision

For those of you who have never coded in Python, here are a few interesting resources to get started.

• TutorialPoint
• Programiz

Jupyter notebooks and lab

In order to ease following the exercises along with the course, we will be relying on Jupyter Notebooks. If you have never used a notebook before, we recommend that you look at their website to understand the concept. Here we also provide the instructions to install Jupyter Lab which is a more integrative version of notebooks. You can install it on your computer as follows (if you use pip)

pip install jupyterlab

Then, once installed, you can go to the folder where you cloned this repository, and type in

jupyter lab

In order to test our algorithms on audio and music data, we will work with several datasets that should be downloaded on your local computer first from this link

Unzip the file and place the 00_Datasets folder along with the other code folders For the first parts of the tutorial, we will mostly rely solely on the classification dataset. In order to facilitate the interactions, we provide the function importDataset that will allow to import all audio datasets along the tutorials.

function dataStruct = importDataset(classPath, type)
% classPath  : Path to the dataset (string)
% type       : Type of dataset (string: 'classify', 'plain', 'metadata')

% Returns the dataStruct structure with
dataStruct.filenames  % Cell containing the list of audio files
dataStruct.classes    % Vector of indexes assigning each file to a class
dataStruct.classNames % Cell of class names

Expected output [Reveal]

  altotrombone 	: 13.
       animals 	: 9.
         bells 	: 7.
    cellobowed 	: 47.
        crowds 	: 4.
      laughter 	: 7.
      machines 	: 11.
          oboe 	: 32.
    percussion 	: 99.
 speech.female 	: 35.
   speech.male 	: 17.
     telephone 	: 17.
  tubularbells 	: 20.
   violinbowed 	: 45.
    violinpizz 	: 40.
         water 	: 7.
         total 	: 410.

We will rely on a set of spectral transforms that allow to obtain a more descriptive view over the audio information. As most of these are out of the scope of the machine learning course, we redirect you to a signal processing course proposed by Julius O. Smith.

The following functions to compute various types of transforms are given as part of the basic package, in the 00_Preprocessing folder

In order to perform the various computations, we provide the following function, which performs the different transforms on a complete dataset.

function dataStruct = computeTransforms(dataStruct)
% dataStruct   : Dataset structure with filenames

% Returns the dataStruct structure with
dataStruct.spectrumPower     % Power spectrum (STFT)
dataStruct.spectrumBark      % Spectrum in Bark scale
dataStruct.spectrumMel       % Spectrum in Mel scale
dataStruct.spectrumChroma    % Chroma vectors
dataStruct.spectrumCepstrum  % Cepstrum
dataStruct.spectrumConstantQ % Constant-Q transform

Expected output [Reveal]

As you might have noted from the previous exercice, most spectral transforms have a very high dimensionality, and might not be suited to exhibit the relevant structure of different classes. To that end, we provide a set of functions for computing several spectral features in the 00_Features folder, we redirect interested readers to this exhaustive article on spectral features computation.

Once again, we provide a function to perform the computation of different features on a complete set. Note that for each feature, we compute the temporal evolution in a vector along with the mean and standard deviation of each feature. We only detail the resulting data structure for a single feature (SpectralCentroid).

function dataStruct = computeFeatures(dataStruct)
% dataStruct   : Dataset structure with filenames

% Returns the dataStruct structure with
dataStruct.SpectralCentroid     % Temporal value of a feature
dataStruct.SpectralCentroidMean % Mean value of that feature
dataStruct.SpectralCentroidStd  % Standard deviation

Expected output [Reveal]