{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import numpy, scipy, matplotlib.pyplot as plt, IPython.display, sklearn\n",
    "import essentia, essentia.standard as ess\n",
    "plt.rcParams['figure.figsize'] = (14,5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "[← Back to Index](index.html)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Spectral Features in Essentia"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For classification, we're going to be using new features in our arsenal: spectral moments (centroid, bandwidth, skewness, kurtosis) and other spectral statistics."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "[*Moments*](https://en.wikipedia.org/wiki/Moment_(mathematics) is a term used in physics and statistics. There are raw moments and central moments. \n",
    "\n",
    "You are probably already familiar with two examples of moments: mean and variance. The first raw moment is known as the mean. The second central moment is known as the variance."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## `essentia.standard.Centroid`"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To compute the spectral centroid in Essentia, we will use [`essentia.standard.Centroid`](http://essentia.upf.edu/documentation/reference/std_Centroid.html):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.498104333878\n"
     ]
    }
   ],
   "source": [
    "spectrum = ess.Spectrum()\n",
    "centroid = ess.Centroid()\n",
    "x = essentia.array(scipy.randn(1024))\n",
    "X = spectrum(x)\n",
    "spectral_centroid = centroid(X)\n",
    "print spectral_centroid"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This value is normalized between 0 and 1. If 0, then the centroid is at zero. If 1, then the centroid is all the way to the \"right\", i.e., equal to `fs/2`, the Nyquist frequency, or the highest frequency a digital signal can possibly have."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here is a sanity check:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.49810430034108322"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sum((X/sum(X))*numpy.linspace(0, 1, len(X)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## `essentia.standard.CentralMoments`"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The first step to computing the other three spectral moments (spread, skewness, and kurtosis) is to compute the central moments of a spectrum:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[  1.00000000e+00   0.00000000e+00   8.12825486e-02  -3.57219629e-04\n",
      "   1.20308148e-02]\n"
     ]
    }
   ],
   "source": [
    "central_moments = ess.CentralMoments()\n",
    "print central_moments(X)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## `essentia.standard.DistributionShape`"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To compute the spectral spread, skewness, and kurtosis, we use [`essentia.standard.DistributionShape`](http://essentia.upf.edu/documentation/reference/std_DistributionShape.html):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(0.08128254860639572, -0.01541485171765089, -1.1790399551391602)\n"
     ]
    }
   ],
   "source": [
    "distributionshape = ess.DistributionShape()\n",
    "spectral_moments = distributionshape(central_moments(X))\n",
    "print spectral_moments"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "[← Back to Index](index.html)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}