musicfeatures.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from __future__ import print_function\n",
    "from pathlib import Path"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# We'll need numpy for some mathematical operations\n",
    "import numpy as np\n",
    "\n",
    "\n",
    "# matplotlib for displaying the output\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.style as ms\n",
    "ms.use('seaborn-muted')\n",
    "%matplotlib inline\n",
    "\n",
    "\n",
    "# and IPython.display for audio output\n",
    "import IPython.display\n",
    "\n",
    "\n",
    "# Librosa for audio\n",
    "import librosa\n",
    "# And the display module for visualization\n",
    "import librosa.display\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Imports\n",
    "# pandas\n",
    "import pandas as pd\n",
    "from pandas import Series,DataFrame\n",
    "# numpy, matplotlib, seaborn\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from datetime import datetime\n",
    "%matplotlib inline\n",
    "np.set_printoptions(precision=5, suppress=True)  # suppress scientific float notation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "mp3dir = 'piano_first_thirty_sec'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#colnames = ['composer','description', 'symphony', 'file', 'part', 'tempo', 'bpm']\n",
    "musicdf = pd.read_csv('./musicdf.csv', index_col=0, encoding='latin-1')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def load(audio_path):\n",
    "    audio_path = '.\\\\'+ mp3dir + '\\\\' + audio_path\n",
    "    my_file = Path(audio_path)\n",
    "    if my_file.is_file():\n",
    "        return librosa.load(audio_path, sr=None)\n",
    "    else:\n",
    "        return None, None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def getbpm(audio_path):\n",
    "    y, sr = load(audio_path)\n",
    "    onset_env = librosa.onset.onset_strength(y, sr=sr)\n",
    "    return librosa.beat.tempo(onset_envelope=onset_env, sr=sr)[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def getDuration(audio_path):\n",
    "    y, sr = load(audio_path)\n",
    "    return librosa.core.get_duration(y,sr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PolyCollection at 0x254644405c0>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEKCAYAAAAfGVI8AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x25464350748>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "x, sr= load(musicdf.iloc[0].file)\n",
    "librosa.display.waveplot(x, sr=sr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def getZeroCrossings(audio_path):\n",
    "    y, sr = load(audio_path)\n",
    "    return sum(librosa.zero_crossings(y, pad=False, threshold=0.001))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "## http://musicinformationretrieval.com/spectral_features.html\n",
    "## http://cs229.stanford.edu/proj2016/report/NiShiWugofski_FinalReport.pdf\n",
    "def getAvgSpectralCentroid(audio_path):\n",
    "    y, sr = load(audio_path)\n",
    "    spectral_centroids = librosa.feature.spectral_centroid(y+0.01, sr=sr)[0]\n",
    "    return spectral_centroids.mean()\n",
    "\n",
    "def getAvgSpectralBW(audio_path):\n",
    "    y, sr = load(audio_path)\n",
    "    spectral_bandwidth_2 = librosa.feature.spectral_bandwidth(y+0.01, sr=sr)[0]\n",
    "    return spectral_bandwidth_2.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "## http://musicinformationretrieval.com/chroma.html\n",
    "hop_length = 512\n",
    "def chroma(audio_path):\n",
    "    y, sr = load(audio_path)\n",
    "    chromagram = librosa.feature.chroma_cens(y, sr=sr, hop_length=hop_length)\n",
    "    return pd.Series({\n",
    "              \"Chroma_C\":chromagram[0].mean(), \"Chroma_C#\":chromagram[1].mean(),\n",
    "              \"Chroma_D\":chromagram[2].mean(), \"Chroma_D#\":chromagram[3].mean(),\n",
    "              \"Chroma_E\":chromagram[4].mean(), \"Chroma_F\":chromagram[5].mean(),\n",
    "              \"Chroma_F#\":chromagram[6].mean(), \"Chroma_G\":chromagram[7].mean(),\n",
    "              \"Chroma_G#\":chromagram[8].mean(), \"Chroma_A\":chromagram[9].mean(),\n",
    "              \"Chroma_A#\":chromagram[10].mean(), \"Chroma_B\":chromagram[11].mean(),\n",
    "              \"Chroma_SD_C\":chromagram[0].std(), \"Chroma_SD_C#\":chromagram[1].std(),\n",
    "              \"Chroma_SD_D\":chromagram[2].std(), \"Chroma_SD_D#\":chromagram[3].std(),\n",
    "              \"Chroma_SD_E\":chromagram[4].std(), \"Chroma_SD_F\":chromagram[5].std(),\n",
    "              \"Chroma_SD_F#\":chromagram[6].std(), \"Chroma_SD_G\":chromagram[7].std(),\n",
    "              \"Chroma_SD_G#\":chromagram[8].std(), \"Chroma_SD_A\":chromagram[9].std(),\n",
    "              \"Chroma_SD_A#\":chromagram[10].std(), \"Chroma_SD_B\":chromagram[11].std()})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def mfcc(audio_path):\n",
    "    y, sr = load(audio_path)\n",
    "    mfcc = librosa.feature.mfcc(y, sr=sr, n_mfcc=12)\n",
    "    return pd.Series({\n",
    "              \"MFCC_0\":mfcc[0].mean(), \"MFCC_1\":mfcc[1].mean(),\n",
    "              \"MFCC_2\":mfcc[2].mean(), \"MFCC_3\":mfcc[3].mean(),\n",
    "              \"MFCC_4\":mfcc[4].mean(), \"MFCC_5\":mfcc[5].mean(),\n",
    "              \"MFCC_6\":mfcc[6].mean(), \"MFCC_7\":mfcc[7].mean(),\n",
    "              \"MFCC_8\":mfcc[8].mean(), \"MFCC_9\":mfcc[9].mean(),\n",
    "              \"MFCC_10\":mfcc[10].mean(), \"MFCC_11\":mfcc[11].mean(),\n",
    "              \"MFCC_SD_0\":mfcc[0].std(), \"MFCC_SD_1\":mfcc[1].std(),\n",
    "              \"MFCC_SD_2\":mfcc[2].std(), \"MFCC_SD_3\":mfcc[3].std(),\n",
    "              \"MFCC_SD_4\":mfcc[4].std(), \"MFCC_SD_5\":mfcc[5].std(),\n",
    "              \"MFCC_SD_6\":mfcc[6].std(), \"MFCC_SD_7\":mfcc[7].std(),\n",
    "              \"MFCC_SD_8\":mfcc[8].std(), \"MFCC_SD_9\":mfcc[9].std(),\n",
    "              \"MFCC_SD_10\":mfcc[10].std(), \"MFCC_SD_11\":mfcc[11].std()\n",
    "    })    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "musicdf['bpm'] = musicdf['file'].apply(getbpm)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "musicdf['duration'] = musicdf['file'].apply(getDuration)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "musicdf['zero_crossings'] = musicdf['file'].apply(getZeroCrossings)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "musicdf['spectral_centroid_avg'] =  musicdf['file'].apply(getAvgSpectralCentroid)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "musicdf['spectral_bw_avg'] =  musicdf['file'].apply(getAvgSpectralBW)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "musicdf = pd.concat([musicdf, musicdf['file'].apply(chroma)], axis=1)\n",
    "#musicdf.update(musicdf['file'].apply(chroma))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "musicdf = pd.concat([musicdf, musicdf['file'].apply(mfcc)], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "musicdf.to_csv('musicdf30.csv', encoding='latin-1')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(61, 59)"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "musicdf.shape"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}