{
 "metadata": {
  "name": ""
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%pylab --no-import-all inline\n",
      "execfile('load_data.py')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Populating the interactive namespace from numpy and matplotlib\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stderr",
       "text": [
        "WARNING: pylab import has clobbered these variables: ['plt']\n",
        "`%pylab --no-import-all` prevents importing * from pylab and numpy\n"
       ]
      }
     ],
     "prompt_number": 228
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from IPython.display import HTML\n",
      "def html_table( tbl ):\n",
      "    s = \"<table>\"\n",
      "    maxwidth = max( map( len, tbl) )\n",
      "    for row in tbl:\n",
      "        if len(row)==1:\n",
      "            s += \"<tr><td colspan=%d>\" % maxwidth\n",
      "            s += row[0]\n",
      "            s += \"</td></tr>\"\n",
      "        else:\n",
      "            s += \"<tr><td>\" + \"</td><td>\".join( row ) + \"</td></tr>\"\n",
      "    s += \"</table>\"\n",
      "    return s"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 229
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# Basic info\n",
      "tbl = [ [ 'Set',      'Spoken sentences', 'Spoken phonemes', 'Speakers' ] ]\n",
      "\n",
      "for desc, data in zip( ('Training', 'Validation', 'Test'), (train,valid,test) ):\n",
      "    tbl.append( [ desc, str(data.number_of_recorded_sentences()),\n",
      "                              str(data.number_of_recorded_phonemes()),\n",
      "                              str(data.number_of_distinct_speakers()) ] )\n",
      "assert( len(np.intersect1d( train.spkr, test.spkr))==0 )\n",
      "assert( len(np.intersect1d( valid.spkr, test.spkr))==0 )\n",
      "assert( len(np.intersect1d( train.spkr, valid.spkr))==0 )\n",
      "tbl.append( [ 'Train, valid and test set have no speakers in common' ] )\n",
      "\n",
      "HTML( html_table( tbl ) )\n",
      "\n",
      "\n"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<table><tr><td>Set</td><td>Spoken sentences</td><td>Spoken phonemes</td><td>Speakers</td></tr><tr><td>Training</td><td>4120</td><td>158084</td><td>412</td></tr><tr><td>Validation</td><td>500</td><td>18996</td><td>50</td></tr><tr><td>Test</td><td>1680</td><td>64145</td><td>168</td></tr><tr><td colspan=4>Train, valid and test set have no speakers in common</td></tr></table>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 230,
       "text": [
        "<IPython.core.display.HTML at 0x34adedd0>"
       ]
      }
     ],
     "prompt_number": 230
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# List of all phonemes\n",
      "tbl = [ ['List of all 61 phonemes' ]]\n",
      "tbl.append( phonemes[0:31] )\n",
      "tbl.append( phonemes[31:] )\n",
      "\n",
      "print  html_table(tbl) \n",
      "\n",
      "HTML( html_table(tbl) )"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "<table><tr><td colspan=31>List of all 61 phonemes</td></tr><tr><td>iy</td><td>ch</td><td>el</td><td>tcl</td><td>h#</td><td>pcl</td><td>bcl</td><td>zh</td><td>th</td><td>dh</td><td>kcl</td><td>hv</td><td>hh</td><td>dx</td><td>ax-h</td><td>em</td><td>d</td><td>b</td><td>ux</td><td>f</td><td>uw</td><td>l</td><td>n</td><td>p</td><td>r</td><td>t</td><td>v</td><td>z</td><td>aa</td><td>ix</td><td>en</td></tr><tr><td>ae</td><td>eh</td><td>ah</td><td>ao</td><td>ih</td><td>ey</td><td>aw</td><td>ay</td><td>ax</td><td>er</td><td>pau</td><td>eng</td><td>gcl</td><td>ng</td><td>nx</td><td>uh</td><td>dcl</td><td>w</td><td>ow</td><td>jh</td><td>axr</td><td>g</td><td>k</td><td>m</td><td>q</td><td>s</td><td>sh</td><td>oy</td><td>epi</td><td>y</td></tr></table>\n"
       ]
      },
      {
       "html": [
        "<table><tr><td colspan=31>List of all 61 phonemes</td></tr><tr><td>iy</td><td>ch</td><td>el</td><td>tcl</td><td>h#</td><td>pcl</td><td>bcl</td><td>zh</td><td>th</td><td>dh</td><td>kcl</td><td>hv</td><td>hh</td><td>dx</td><td>ax-h</td><td>em</td><td>d</td><td>b</td><td>ux</td><td>f</td><td>uw</td><td>l</td><td>n</td><td>p</td><td>r</td><td>t</td><td>v</td><td>z</td><td>aa</td><td>ix</td><td>en</td></tr><tr><td>ae</td><td>eh</td><td>ah</td><td>ao</td><td>ih</td><td>ey</td><td>aw</td><td>ay</td><td>ax</td><td>er</td><td>pau</td><td>eng</td><td>gcl</td><td>ng</td><td>nx</td><td>uh</td><td>dcl</td><td>w</td><td>ow</td><td>jh</td><td>axr</td><td>g</td><td>k</td><td>m</td><td>q</td><td>s</td><td>sh</td><td>oy</td><td>epi</td><td>y</td></tr></table>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 238,
       "text": [
        "<IPython.core.display.HTML at 0x26c1890>"
       ]
      }
     ],
     "prompt_number": 238
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "phoneme_num_utterances = []\n",
      "rows = []\n",
      "for i in range(len(phonemes)):\n",
      "    num_utterances = sum(train.phn[:,2]==i)\n",
      "    row = (num_utterances,phonemes[i])\n",
      "    rows.append( row )\n",
      "\n",
      "rows.sort(reverse=True)\n",
      "\n",
      "tbl = [ ['Number of occurences of each phoneme in training set' ] ]\n",
      "tbl.append( ['Phoneme', 'Occurences' ] )\n",
      "tbl = tbl + map( lambda a: [a[1], str(a[0])], rows )\n",
      "HTML( html_table(tbl) )"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<table><tr><td colspan=2>Number of occurences of each phoneme in training set</td></tr><tr><td>Phoneme</td><td>Occurences</td></tr><tr><td>h#</td><td>8240</td></tr><tr><td>ix</td><td>7731</td></tr><tr><td>s</td><td>6680</td></tr><tr><td>n</td><td>6315</td></tr><tr><td>iy</td><td>6219</td></tr><tr><td>tcl</td><td>5951</td></tr><tr><td>r</td><td>5855</td></tr><tr><td>kcl</td><td>5230</td></tr><tr><td>l</td><td>5160</td></tr><tr><td>ih</td><td>4543</td></tr><tr><td>dcl</td><td>4400</td></tr><tr><td>k</td><td>4338</td></tr><tr><td>t</td><td>3925</td></tr><tr><td>ae</td><td>3571</td></tr><tr><td>m</td><td>3518</td></tr><tr><td>eh</td><td>3448</td></tr><tr><td>z</td><td>3367</td></tr><tr><td>q</td><td>3215</td></tr><tr><td>ax</td><td>3201</td></tr><tr><td>d</td><td>3157</td></tr><tr><td>axr</td><td>3013</td></tr><tr><td>w</td><td>2789</td></tr><tr><td>aa</td><td>2757</td></tr><tr><td>ao</td><td>2650</td></tr><tr><td>dh</td><td>2501</td></tr><tr><td>dx</td><td>2415</td></tr><tr><td>pcl</td><td>2380</td></tr><tr><td>p</td><td>2341</td></tr><tr><td>ay</td><td>2123</td></tr><tr><td>ah</td><td>2049</td></tr><tr><td>ey</td><td>2041</td></tr><tr><td>sh</td><td>2021</td></tr><tr><td>f</td><td>1986</td></tr><tr><td>gcl</td><td>1973</td></tr><tr><td>b</td><td>1930</td></tr><tr><td>ow</td><td>1906</td></tr><tr><td>er</td><td>1808</td></tr><tr><td>g</td><td>1786</td></tr><tr><td>v</td><td>1771</td></tr><tr><td>bcl</td><td>1690</td></tr><tr><td>ux</td><td>1663</td></tr><tr><td>y</td><td>1510</td></tr><tr><td>epi</td><td>1320</td></tr><tr><td>ng</td><td>1186</td></tr><tr><td>jh</td><td>1077</td></tr><tr><td>hv</td><td>1022</td></tr><tr><td>pau</td><td>868</td></tr><tr><td>el</td><td>864</td></tr><tr><td>hh</td><td>854</td></tr><tr><td>nx</td><td>852</td></tr><tr><td>ch</td><td>714</td></tr><tr><td>th</td><td>675</td></tr><tr><td>aw</td><td>643</td></tr><tr><td>en</td><td>639</td></tr><tr><td>oy</td><td>602</td></tr><tr><td>uw</td><td>506</td></tr><tr><td>uh</td><td>483</td></tr><tr><td>ax-h</td><td>343</td></tr><tr><td>zh</td><td>128</td></tr><tr><td>em</td><td>108</td></tr><tr><td>eng</td><td>33</td></tr></table>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 140,
       "text": [
        "<IPython.core.display.HTML at 0x435b2910>"
       ]
      }
     ],
     "prompt_number": 140
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import operator\n",
      "def dict_count_add( dic, key ):\n",
      "    if key in dic:\n",
      "        dic[key] = dic[key] + 1\n",
      "    else:\n",
      "        dic[key] = 1\n",
      "\n",
      "\n",
      "output = \"Number of recordings of each sentence:<br>\"\n",
      "counts = {}\n",
      "sorted_counts = {}\n",
      "# Number of unique senences\n",
      "for desc,dataset in (('training set', train), ('validation set', valid), ('test set', test)):\n",
      "    count = {}\n",
      "    for i in range(len(dataset.x_raw)):\n",
      "        s = tuple(dataset.sentence_idx_to_word_nums(i))   \n",
      "        dict_count_add( count, s )\n",
      "    tbl = [ [desc] ]\n",
      "    tbl.append( ['# Sentences that have', '# recordings'] )\n",
      "    for a in range(max(count.values())+1,0,-1):\n",
      "        if sum(array(count.values())==a)!=0:\n",
      "            tbl.append( [ str( sum(array(count.values())==a) ), str(a) ] )\n",
      "    tbl.append( [ str(len(count.values())), '<- total number of unique sentences' ] )\n",
      "    output += html_table(tbl)\n",
      "    sorted_count = sorted(count.iteritems(), key=operator.itemgetter(1),reverse=True)\n",
      "    tbl = [ [\"Two most comment sentences in %s\" % desc ] ]\n",
      "    tbl.append( [\" \".join( word_num_to_word_str(list(sorted_count[0][0]) ) ) ])\n",
      "    tbl.append( [\" \".join( word_num_to_word_str(list(sorted_count[1][0]) ) ) ])\n",
      "    output += html_table(tbl) + \"<br>\"\n",
      "    \n",
      "    counts[dataset] = count\n",
      "    sorted_counts[dataset] = sorted_count\n",
      "\n",
      "\n",
      "assert( len(np.intersect1d(counts[train].keys(),counts[test].keys()))==2 )\n",
      "assert( len(np.intersect1d(counts[valid].keys(),counts[test].keys()))==2 )\n",
      "output += \"(Training+validiation) and test set have only the two \\\"special\\\" sentences in common.<br>\"\n",
      "output += \"Training and validiation sets have a further \"\n",
      "output += str(len(np.intersect1d(counts[train].keys(),counts[valid].keys()))-2) + \" sentences in common\"\n",
      "\n",
      "HTML(output)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "Number of recordings of each sentence:<br><table><tr><td colspan=2>training set</td></tr><tr><td># Sentences that have</td><td># recordings</td></tr><tr><td>2</td><td>412</td></tr><tr><td>136</td><td>7</td></tr><tr><td>144</td><td>6</td></tr><tr><td>36</td><td>5</td></tr><tr><td>8</td><td>4</td></tr><tr><td>6</td><td>3</td></tr><tr><td>1</td><td>2</td></tr><tr><td>1248</td><td>1</td></tr><tr><td>1581</td><td><- total number of unique sentences</td></tr></table><table><tr><td colspan=1>Two most comment sentences in training set</td></tr><tr><td colspan=1>she had your dark suit in greasy wash water all year</td></tr><tr><td colspan=1>don't ask me to carry an oily rag like that</td></tr></table><br><table><tr><td colspan=2>validation set</td></tr><tr><td># Sentences that have</td><td># recordings</td></tr><tr><td>2</td><td>50</td></tr><tr><td>6</td><td>4</td></tr><tr><td>4</td><td>3</td></tr><tr><td>35</td><td>2</td></tr><tr><td>294</td><td>1</td></tr><tr><td>341</td><td><- total number of unique sentences</td></tr></table><table><tr><td colspan=1>Two most comment sentences in validation set</td></tr><tr><td colspan=1>she had your dark suit in greasy wash water all year</td></tr><tr><td colspan=1>don't ask me to carry an oily rag like that</td></tr></table><br><table><tr><td colspan=2>test set</td></tr><tr><td># Sentences that have</td><td># recordings</td></tr><tr><td>2</td><td>168</td></tr><tr><td>114</td><td>7</td></tr><tr><td>5</td><td>6</td></tr><tr><td>1</td><td>5</td></tr><tr><td>1</td><td>2</td></tr><tr><td>509</td><td>1</td></tr><tr><td>632</td><td><- total number of unique sentences</td></tr></table><table><tr><td colspan=1>Two most comment sentences in test set</td></tr><tr><td colspan=1>she had your dark suit in greasy wash water all year</td></tr><tr><td colspan=1>don't ask me to carry an oily rag like that</td></tr></table><br>(Training+validiation) and test set have only the two \"special\" sentences in common.<br>Training and validiation sets have a further 187 sentences in common"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 162,
       "text": [
        "<IPython.core.display.HTML at 0x4353ef50>"
       ]
      }
     ],
     "prompt_number": 162
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "print \"Number of phonetic variations of 10 most common sentences\"\n",
      "sorted_count = sorted_counts[train]\n",
      "# Count phonetic variations of each sentence\n",
      "for sent_wrds,count in sorted_count[0:10]:\n",
      "    ct_phn = {}\n",
      "    for idx in range(len(train.x_raw)):\n",
      "        if tuple(train.sentence_idx_to_word_nums(idx))==sent_wrds:\n",
      "            phns = tuple(train.sentence_idx_to_phoneme_nums(idx))\n",
      "            dict_count_add( ct_phn, phns )\n",
      "    print len(ct_phn.values()),\"unique phonetic variations out of\",count,\"recordings\"\n",
      "\n",
      "print \"Some phonetic variations of most common sentences\"\n",
      "print \"Sentence:\", \" \".join( word_num_to_word_str(list(sorted_count[0][0])) )\n",
      "to_show = 2;\n",
      "for i in range(len(train.x_raw)):\n",
      "    if tuple(train.sentence_idx_to_word_nums(i))==sorted_count[0][0]:\n",
      "        print \" \".join( train.sentence_idx_to_phoneme_strs(i) )\n",
      "        to_show -= 1\n",
      "        if to_show==0: break"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Number of phonetic variations of 10 most common sentences\n",
        "410"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " unique phonetic variations out of 412 recordings\n",
        "400"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " unique phonetic variations out of 412 recordings\n",
        "7"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " unique phonetic variations out of 7 recordings\n",
        "7"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " unique phonetic variations out of 7 recordings\n",
        "7"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " unique phonetic variations out of 7 recordings\n",
        "7"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " unique phonetic variations out of 7 recordings\n",
        "7"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " unique phonetic variations out of 7 recordings\n",
        "7"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " unique phonetic variations out of 7 recordings\n",
        "7"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " unique phonetic variations out of 7 recordings\n",
        "7"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " unique phonetic variations out of 7 recordings\n",
        "Some phonetic variations of most common sentences\n",
        "Sentence: she had your dark suit in greasy wash water all year\n",
        "h# sh iy hv ae dcl jh axr dcl d aa r kcl k s ux dx ix ng gcl g r iy z iy w aa sh epi w ao dx axr q ao l y ih axr h#\n",
        "h# s iy eh dcl d axr dcl d aa r kcl s ux q en gcl g r iy s ix w aa sh epi w ao dx er ao l y ih r h#\n"
       ]
      }
     ],
     "prompt_number": 164
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# Lengths of phonemes\n",
      "lengths = []\n",
      "phoneme_lengths = {}\n",
      "for phn in phonemes: phoneme_lengths[phn] = []\n",
      "for idx in range(train.number_of_recorded_phonemes()):\n",
      "    phoneme = train.phoneme_idx_to_phoneme_str(idx)\n",
      "    start, end = train.phoneme_idx_to_offsets(idx)\n",
      "    length = (end-start)/16000.0*1000.0 #Count in milliseconds\n",
      "    phoneme_lengths[phoneme].append( (length,idx) )\n",
      "\n",
      "def first_element( l ): # Get the first element of each tuple in a list of tuples\n",
      "    return map( lambda x: x[0], l )\n",
      "\n",
      "medians = map( lambda (key,val): (numpy.median( first_element( val )),key), phoneme_lengths.iteritems() ) \n",
      "medians.sort(reverse=True)\n",
      "\n",
      "import pylab as plt\n",
      "plt.figure( figsize=(30,5) )\n",
      "axis([0, 62, 0, 500 ])\n",
      "plt.boxplot( map( lambda x: first_element( phoneme_lengths[x[1]]), medians), vert=True )\n",
      "plt.xticks(range(1,62), map( lambda x: x[1], medians) )\n",
      "plt.xlabel('Phoneme')\n",
      "plt.ylabel('Length (ms)')\n",
      "\n",
      "print \"longest and shortest duration of h# phoneme:\", min(phoneme_lengths['h#']), max(phoneme_lengths['h#'])\n",
      "print \"longest and shortest duration of pau phoneme:\", min(phoneme_lengths['pau']), max(phoneme_lengths['pau'])\n",
      "\n",
      "plt.show()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "longest and shortest duration of h# phoneme: (2.0, 98932) (2996.875, 156710)\n",
        "longest and shortest duration of pau phoneme: (22.5625, 29185) (652.5625, 3828)\n"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAABr0AAAFECAYAAACaiPUWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAASUVORK5CYII=\n",
       "text": [
        "<matplotlib.figure.Figure at 0x143c9b90>"
       ]
      }
     ],
     "prompt_number": 218
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [],
     "language": "python",
     "metadata": {},
     "outputs": []
    }
   ],
   "metadata": {}
  }
 ]
}