Skip to content

GitLab

Switch branch/tag
Find file Normal viewHistoryPermalink
Run_ESTER1_2.ipynb 93 KB
EditWeb IDE
Newer Older
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
1 2 3 4 
{
 "cells": [
  {
   "cell_type": "code",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
5 
   "execution_count": 17,
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
6 7 8 9 10 11 12 
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    ". path.sh\n",
    ". cmd.sh\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
13 14 15 
    "idata_kaldi=data_last_ffmpeg\n",
    "exp_kaldi=exp_last_ffmpeg\n",
    "#model_tri_sat_ESTER=exp-ESTER-V4/monoAll_1_2/tri1_10K_100K_ESTER_All/tri1_15K_200K_ESTER_All/tri2_20K_300K_ESTER_All/tri1_SAT_50K_400K/tri1_SAT_70K_500K"
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
16 17 18 19 20 21 22 23 24 25 26 
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Prepare ESTER 2"
   ]
  },
  {
   "cell_type": "code",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
27 28 29 30 
   "execution_count": 18,
   "metadata": {
    "scrolled": true
   },
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
31 32 33 34 35 
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 
      "prepare /fast/LINAGORA/Corpus/database/Corpus/ESTER/DGA/Phase1/data\n",
      "utils/data/get_utt2dur.sh: working out data_last_ffmpeg/data/utt2dur from data_last_ffmpeg/data/segments\n",
      "utils/data/get_utt2dur.sh: computed data_last_ffmpeg/data/utt2dur\n",
      "utils/validate_data_dir.sh: Successfully validated data-directory data_last_ffmpeg/data\n",
      "Successfully prepared data in data_last_ffmpeg/data..\n",
      "prepare /fast/LINAGORA/Corpus/database/Corpus/ESTER/DGA/Phase2/DATA2\n",
      "utils/data/get_utt2dur.sh: working out data_last_ffmpeg/DATA2/utt2dur from data_last_ffmpeg/DATA2/segments\n",
      "utils/data/get_utt2dur.sh: computed data_last_ffmpeg/DATA2/utt2dur\n",
      "utils/validate_data_dir.sh: Successfully validated data-directory data_last_ffmpeg/DATA2\n",
      "Successfully prepared data in data_last_ffmpeg/DATA2..\n",
      "prepare /fast/LINAGORA/Corpus/database/Corpus/ESTER/DGA/Eval2005/DATA\n",
      "utils/data/get_utt2dur.sh: working out data_last_ffmpeg/DATA/utt2dur from data_last_ffmpeg/DATA/segments\n",
      "utils/data/get_utt2dur.sh: computed data_last_ffmpeg/DATA/utt2dur\n",
      "utils/validate_data_dir.sh: Successfully validated data-directory data_last_ffmpeg/DATA\n",
      "Successfully prepared data in data_last_ffmpeg/DATA..\n",
      "prepare /fast/LINAGORA/Corpus/database/Corpus/ESTER2/corpus//train\n",
      "utils/data/get_utt2dur.sh: working out data_last_ffmpeg/train/utt2dur from data_last_ffmpeg/train/segments\n",
      "utils/data/get_utt2dur.sh: computed data_last_ffmpeg/train/utt2dur\n",
      "utils/validate_data_dir.sh: Successfully validated data-directory data_last_ffmpeg/train\n",
      "Successfully prepared data in data_last_ffmpeg/train..\n"
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
56 57 58 59 
     ]
    }
   ],
   "source": [
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
60 61 62 63 64 65 
    "data_ESTER2=/fast/LINAGORA/Corpus/database/Corpus/ESTER2/corpus/\n",
    "data_ESTER_Phase1=/fast/LINAGORA/Corpus/database/Corpus/ESTER/DGA/Phase1\n",
    "data_ESTER_Phase2=/fast/LINAGORA/Corpus/database/Corpus/ESTER/DGA/Phase2\n",
    "data_ESTER_EVAL=/fast/LINAGORA/Corpus/database/Corpus/ESTER/DGA/Eval2005\n",
    "#idata_kaldi=data_last\n",
    "for part in $data_ESTER_Phase1/data $data_ESTER_Phase2/DATA2 $data_ESTER_EVAL/DATA $data_ESTER2/train; do\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
66 67 68 69 
    "  # use underscore-separated names in data directories.\n",
    "  echo \"prepare $part\"\n",
    "  #local/data_prepTCOF.sh $data/$part $idata_kaldi/$part\n",
    "  # probleme event (URL:)\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
70 71 
    "  folder_name=$(basename $part)\n",
    "  local/data_prepESTER.sh $part $idata_kaldi/$folder_name\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
72 73 74 75 76 
    "done"
   ]
  },
  {
   "cell_type": "code",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
77 78 79 80 
   "execution_count": 19,
   "metadata": {
    "scrolled": true
   },
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
81 82 83 84 85 
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 
      "steps/make_mfcc.sh --cmd run.pl --mem 64G --nj 32 data_last_ffmpeg/data exp_last_ffmpeg/make_mfcc/data mfcc\n",
      "utils/validate_data_dir.sh: Successfully validated data-directory data_last_ffmpeg/data\n",
      "steps/make_mfcc.sh [info]: segments file exists: using that.\n",
      "Succeeded creating MFCC features for data\n",
      "steps/compute_cmvn_stats.sh data_last_ffmpeg/data exp_last_ffmpeg/make_mfcc/data mfcc\n",
      "Succeeded creating CMVN stats for data\n",
      "fix_data_dir.sh: kept all 35574 utterances.\n",
      "fix_data_dir.sh: old files are kept in data_last_ffmpeg/data/.backup\n",
      "steps/make_mfcc.sh --cmd run.pl --mem 64G --nj 32 data_last_ffmpeg/DATA2 exp_last_ffmpeg/make_mfcc/DATA2 mfcc\n",
      "utils/validate_data_dir.sh: Successfully validated data-directory data_last_ffmpeg/DATA2\n",
      "steps/make_mfcc.sh [info]: segments file exists: using that.\n",
      "It seems not all of the feature files were successfully processed (56956 != 57341);\n",
      "consider using utils/fix_data_dir.sh data_last_ffmpeg/DATA2\n",
      "Succeeded creating MFCC features for DATA2\n",
      "steps/compute_cmvn_stats.sh data_last_ffmpeg/DATA2 exp_last_ffmpeg/make_mfcc/DATA2 mfcc\n",
      "steps/compute_cmvn_stats.sh: warning: it seems not all of the speakers got cmvn stats (1555 != 1558);\n",
      "Succeeded creating CMVN stats for DATA2\n",
      "utils/fix_data_dir.sh: filtered /tmp/kaldi.A3Y6/speakers from 1558 to 1555 lines based on filter data_last_ffmpeg/DATA2/cmvn.scp.\n",
      "utils/fix_data_dir.sh: filtered data_last_ffmpeg/DATA2/spk2utt from 1558 to 1555 lines based on filter /tmp/kaldi.A3Y6/speakers.\n",
      "utils/fix_data_dir.sh: filtered data_last_ffmpeg/DATA2/spk2gender from 1558 to 1555 lines based on filter /tmp/kaldi.A3Y6/speakers.\n",
      "fix_data_dir.sh: kept 56956 utterances out of 56957\n",
      "utils/fix_data_dir.sh: filtered data_last_ffmpeg/DATA2/wav.scp from 126 to 124 lines based on filter /tmp/kaldi.A3Y6/recordings.\n",
      "fix_data_dir.sh: old files are kept in data_last_ffmpeg/DATA2/.backup\n",
      "steps/make_mfcc.sh --cmd run.pl --mem 64G --nj 32 data_last_ffmpeg/DATA exp_last_ffmpeg/make_mfcc/DATA mfcc\n",
      "utils/validate_data_dir.sh: Successfully validated data-directory data_last_ffmpeg/DATA\n",
      "steps/make_mfcc.sh [info]: segments file exists: using that.\n",
      "Succeeded creating MFCC features for DATA\n",
      "steps/compute_cmvn_stats.sh data_last_ffmpeg/DATA exp_last_ffmpeg/make_mfcc/DATA mfcc\n",
      "Succeeded creating CMVN stats for DATA\n",
      "fix_data_dir.sh: kept all 10486 utterances.\n",
      "fix_data_dir.sh: old files are kept in data_last_ffmpeg/DATA/.backup\n",
      "steps/make_mfcc.sh --cmd run.pl --mem 64G --nj 32 data_last_ffmpeg/train exp_last_ffmpeg/make_mfcc/train mfcc\n",
      "utils/validate_data_dir.sh: Successfully validated data-directory data_last_ffmpeg/train\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
119 
      "steps/make_mfcc.sh [info]: segments file exists: using that.\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
120 121 122 123 124 125 
      "run.pl: 1 / 32 failed, log is in exp_last_ffmpeg/make_mfcc/train/make_mfcc_train.*.log\n",
      "steps/compute_cmvn_stats.sh data_last_ffmpeg/train exp_last_ffmpeg/make_mfcc/train mfcc\n",
      "make_cmvn.sh: no such file data_last_ffmpeg/train/feats.scp\n",
      "fix_data_dir.sh: kept all 87296 utterances.\n",
      "fix_data_dir.sh: old files are kept in data_last_ffmpeg/train/.backup\n",
      "utils/combine_data.sh data_last_ffmpeg/ESTER1 data_last_ffmpeg/data data_last_ffmpeg/DATA2\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
126 127 128 129 130 131 132 133 134 135 136 
      "utils/combine_data.sh [info]: not combining utt2uniq as it does not exist\n",
      "utils/combine_data.sh: combined segments\n",
      "utils/combine_data.sh: combined utt2spk\n",
      "utils/combine_data.sh [info]: not combining utt2lang as it does not exist\n",
      "utils/combine_data.sh: combined utt2dur\n",
      "utils/combine_data.sh: combined feats.scp\n",
      "utils/combine_data.sh: combined text\n",
      "utils/combine_data.sh: combined cmvn.scp\n",
      "utils/combine_data.sh [info]: not combining reco2file_and_channel as it does not exist\n",
      "utils/combine_data.sh: combined wav.scp\n",
      "utils/combine_data.sh: combined spk2gender\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 
      "fix_data_dir.sh: kept all 92530 utterances.\n",
      "fix_data_dir.sh: old files are kept in data_last_ffmpeg/ESTER1/.backup\n",
      "utils/combine_data.sh data_last_ffmpeg/ESTER1_2 data_last_ffmpeg/ESTER1 data_last_ffmpeg/train\n",
      "utils/combine_data.sh [info]: not combining utt2uniq as it does not exist\n",
      "utils/combine_data.sh: combined segments\n",
      "utils/combine_data.sh: combined utt2spk\n",
      "utils/combine_data.sh [info]: not combining utt2lang as it does not exist\n",
      "utils/combine_data.sh: combined utt2dur\n",
      "utils/combine_data.sh [info]: **not combining feats.scp as it does not exist everywhere**\n",
      "utils/combine_data.sh: combined text\n",
      "utils/combine_data.sh [info]: **not combining cmvn.scp as it does not exist everywhere**\n",
      "utils/combine_data.sh [info]: not combining reco2file_and_channel as it does not exist\n",
      "utils/combine_data.sh: combined wav.scp\n",
      "utils/combine_data.sh: combined spk2gender\n",
      "fix_data_dir.sh: kept all 179826 utterances.\n",
      "fix_data_dir.sh: old files are kept in data_last_ffmpeg/ESTER1_2/.backup\n"
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
153 154 155 156 157 
     ]
    }
   ],
   "source": [
    "# Feature Extraction MFCC:\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
158 
    "#exp_kaldi=exp_last\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
159 
    "mfccdir=mfcc\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
160 
    "for part in data DATA2 DATA train; do\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
161 162 163 164 165 166 167 168 
    "    #MFCC features\n",
    "    steps/make_mfcc.sh --cmd \"$train_cmd\" --nj 32 $idata_kaldi/$part $exp_kaldi/make_mfcc/$part $mfccdir\n",
    "    #MFCC features + Pitch\n",
    "    #steps/make_mfcc_pitch.sh --cmd \"$train_cmd\" --nj 12 $idata_kaldi/$part $exp_kaldi/make_mfcc/$part $mfccdir\n",
    "    steps/compute_cmvn_stats.sh $idata_kaldi/$part $exp_kaldi/make_mfcc/$part $mfccdir\n",
    "    utils/fix_data_dir.sh $idata_kaldi/$part\n",
    "done\n",
    "# Combine data ESTER Phase 1 & Phase 2\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
169 170 
    "utils/combine_data.sh $idata_kaldi/ESTER1 $idata_kaldi/data $idata_kaldi/DATA2\n",
    "utils/combine_data.sh $idata_kaldi/ESTER1_2 $idata_kaldi/ESTER1 $idata_kaldi/train"
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
171 172 173 174 175 176 177 178 179 180 181 
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Build Language model using ESTER 1 & ESTER 2"
   ]
  },
  {
   "cell_type": "code",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
182 183 184 185 
   "execution_count": 13,
   "metadata": {
    "scrolled": true
   },
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
186 187 188 189 190 
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
191 
      "Splitting into 32 parts, to allow for parallel processing ...\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
192 
      "Checking the splits ...\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
193 
      "Performing text normalization (32 jobs) - check data_last_ffmpeg/local_ESTER12/lm/norm/tmp/txt_norm.JOB.log ...\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
194 195 196 
      "Finished OK\n",
      "Selecting the vocabulary (400000 words) ...\n",
      "Making the corpus and the vocabulary ...\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
197 198 199 
      "Word counts saved to 'data_last_ffmpeg/local_ESTER12/lm/word_counts.txt'\n",
      "Vocabulary saved as 'data_last_ffmpeg/local_ESTER12/lm/meeting-vocab.txt'\n",
      "All unique sentences (in sorted order) stored in 'data_last_ffmpeg/local_ESTER12/lm/meeting-lm-norm.txt.gz'\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
200 201 202 203 204 
      "Counting the total number word tokens in the corpus ...\n",
      "There are 2066518 tokens in the corpus\n",
      "Training a 3-gram LM ...\n",
      "This implementation assumes that you have a lot of free RAM(> 12GB) on your machine\n",
      "If that's not the case, consider something like: http://joshua-decoder.org/4.0/large-lms.html\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
205 
      "6,4M\tdata_last_ffmpeg/local_ESTER12/lm/lm_tglarge.arpa.gz\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
206 
      "Creating a 'small' pruned 3-gram LM (threshold: 0.0000003) ...\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
207 208 
      "data_last_ffmpeg/local_ESTER12/lm/lm_tglarge.arpa.gz: line 10: warning: non-zero probability for <unk> in closed-vocabulary LM\n",
      "4,1M\tdata_last_ffmpeg/local_ESTER12/lm/lm_tgsmall.arpa.gz\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
209 
      "Creating a 'medium' pruned 3-gram LM (threshold: 0.0000001) ...\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
210 211 
      "data_last_ffmpeg/local_ESTER12/lm/lm_tglarge.arpa.gz: line 10: warning: non-zero probability for <unk> in closed-vocabulary LM\n",
      "5,4M\tdata_last_ffmpeg/local_ESTER12/lm/lm_tgmed.arpa.gz\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
212 
      "Training a 4-gram LM ...\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
213 
      "7,9M\tdata_last_ffmpeg/local_ESTER12/lm/lm_fglarge.arpa.gz\n"
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
214 215 216 217 
     ]
    }
   ],
   "source": [
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
218 
    "corpus_path=/fast/LINAGORA/Corpus/database\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 
    "LM_train_text=$corpus_path/Textall\n",
    "local/lm/train_lm.sh $LM_train_text \\\n",
    "$idata_kaldi/local_ESTER12/lm/norm/tmp $idata_kaldi/local_ESTER12/lm/norm/norm_texts $idata_kaldi/local_ESTER12/lm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/fast/LINAGORA/tools/LM/lm_tgsphinx.arpa.gz: line 35: warning: non-zero probability for <unk> in closed-vocabulary LM\n",
      "data-ESTER-V4/local_ESTER12/lm/lm_tglarge.arpa.gz: line 10: warning: non-zero probability for <unk> in closed-vocabulary LM\n",
      "BOW numerator for context \"\" is -3.2194e-05 < 0\n",
      "BOW numerator for context \"c' puisque\" is -0.000595476 < 0\n",
      "BOW numerator for context \"c' que\" is -0.000972837 < 0\n",
      "BOW numerator for context \"c' matin\" is -0.000964833 < 0\n",
      "BOW numerator for context \"c' monde\" is -0.000930619 < 0\n",
      "reading 74451 1-grams\n",
      "exp-ESTER-V4/eval_LM_ESTER12/mixed.gz: line 35: warning: non-zero probability for <unk> in closed-vocabulary LM\n",
      "reading 18615477 2-grams\n",
      "reading 23653218 3-grams\n"
     ]
    }
   ],
   "source": [
    "# Evaluate LM\n",
    "testfile=$idata_kaldi/DATA/text_without_noise_tag\n",
    "# LM's\n",
    "LM_LIUM_LARGE=/fast/LINAGORA/tools/LM/lm_tgsphinx.arpa.gz\n",
    "LM_LIUM_SMALL=/fast/LINAGORA/tools/LM/lm_french-small.arpa.gz\n",
    "LM_ESTER_tglarge=$idata_kaldi/local_ESTER12/lm/lm_tglarge.arpa.gz\n",
    "LM_ESTER_fglarge=$idata_kaldi/local_ESTER12/lm/lm_fglarge.arpa.gz\n",
    "# dir eval language model\n",
    "dir_eval_lm=$exp_kaldi/eval_LM_ESTER12\n",
    "mkdir -p $dir_eval_lm\n",
    "\n",
    "# Get Text from kaldi text format\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
260 261 
    "cut -f2- -d' ' < $idata_kaldi/DATA/text | cut -d ' ' -f2- |\\\n",
    "sed -e 's/[ ]\\+/ /g' | sed -e 's/<[^ ][^ ]*>\\|!sil//g' > $testfile\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
262 
    "# Lium 3-gram all\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
263 
    "ngram -lm $LM_LIUM_LARGE -ppl $testfile -debug 2 > $dir_eval_lm/LM_tgLium_full.ppl\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
264 
    "# Lium 3-gram pruné\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
265 
    "ngram -lm $LM_LIUM_SMALL -ppl $testfile -debug 2 > $dir_eval_lm/LM_tgLium_small.ppl\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
266 
    "# ESTER 3-gram\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
267 
    "ngram -lm $LM_ESTER_tglarge -ppl $testfile -debug 2 > $dir_eval_lm/LM_tgESTER.ppl\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
268 
    "# ESTER 4-gram\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
269 
    "ngram -lm $LM_ESTER_fglarge -ppl $testfile -debug 2 > $dir_eval_lm/LM_fgESTER.ppl\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
270 271 
    "#===== Mixe languages\n",
    "# compute best lambda\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
272 
    "compute-best-mix $dir_eval_lm/LM_tgLium_full.ppl $dir_eval_lm/LM_tgESTER.ppl > $dir_eval_lm/best_mix\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
273 274 275 276 277 278 279 280 
    "# mixe languages\n",
    "ngram -lm $LM_LIUM_LARGE -mix-lm $LM_ESTER_tglarge -lambda 0.597881 -write-lm $dir_eval_lm/mixed.gz\n",
    "# compute perplexity\n",
    "ngram -lm $dir_eval_lm/mixed.gz -ppl $testfile -debug 2 > $dir_eval_lm/LM_mixed.ppl"
   ]
  },
  {
   "cell_type": "code",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
281 
   "execution_count": 14,
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " 105023 \n",
      "  52878 aa\n",
      "      1 aaq\n",
      "  49016 ai\n",
      "  20328 an\n",
      "   6620 au\n",
      "  13761 bb\n",
      "   5354 ch\n",
      "  20387 dd\n",
      "  32294 ee\n",
      "  22574 ei\n",
      "   1816 eu\n",
      "  11273 ff\n",
      "   8086 gg\n",
      "   1043 gn\n",
      "  47329 ii\n",
      "   7160 in\n",
      "   8405 jj\n",
      "  29273 kk\n",
      "  31392 ll\n",
      "  21284 mm\n",
      "  21994 nn\n",
      "   2169 oe\n",
      "  10989 on\n",
      "  25795 oo\n",
      "   7341 ou\n",
      "  20469 pp\n",
      "  64183 rr\n",
      "  38533 ss\n",
      "  47040 tt\n",
      "    182 un\n",
      "  13874 uu\n",
      "   1225 uy\n",
      "  11965 vv\n",
      "   3746 ww\n",
      "  16421 yy\n",
      "  28250 zz\n",
      "Downloading and preparing CMUdict\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
328 329 330 331 
      "Autogenerating pronunciations for the words in data_last_ffmpeg/local_ESTER12/dict/g2p/vocab_autogen.* ...\n",
      "12125\n",
      "12125\n",
      "12125 pronunciations autogenerated OK\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
332 
      "Combining the CMUdict pronunciations with the autogenerated ones ...\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
333 
      "Combined lexicon saved to 'data_last_ffmpeg/local_ESTER12/dict/lexicon_raw_nosil.txt'\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
334 
      "Preparing phone lists and clustering questions\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
335 336 337 338 339 
      "4 silence phones saved to: data_last_ffmpeg/local_ESTER12/dict/silence_phones.txt\n",
      "1 optional silence saved to: data_last_ffmpeg/local_ESTER12/dict/optional_silence.txt\n",
      "36 non-silence phones saved to: data_last_ffmpeg/local_ESTER12/dict/nonsilence_phones.txt\n",
      "2 extra triphone clustering-related questions saved to: data_last_ffmpeg/local_ESTER12/dict/extra_questions.txt\n",
      "Lexicon text file saved as: data_last_ffmpeg/local_ESTER12/dict/lexicon.txt\n"
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
340 341 342 343 344 345 346 347 348 349 
     ]
    }
   ],
   "source": [
    "#### Prepare dict: add words which doesn't exist in dictionnary + config files...\n",
    "# print number of phonem used in french\n",
    "dir_repos=/fast/LINAGORA/STT/Thesis_aheba\n",
    "#dir_repos=/home/lingora/Documents/Linagora/kaldi/egs/Linagora/Thesis_aheba\n",
    "cat $dir_repos/cmu_dict/fr.dict | awk '{$1=\"\";print $0}' | tr ' ' '\\n' | sort -b | uniq -c\n",
    "mkdir -p $idata_kaldi/local_ESTER12/dict/cmudict\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
350 
    "cp $dir_repos/cmu_dict/lexicon_new.dict $idata_kaldi/local_ESTER12/dict/fr.dict\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
351 352 353 
    "mkdir -p $idata_kaldi/local_ESTER12/lm/g2p\n",
    "cp $dir_repos/g2p/model-5 $idata_kaldi/local_ESTER12/lm/g2p\n",
    "\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
354 
    "local/prepare_dict.sh --stage 0 --nj 16 --cmd \"$train_cmd\" \\\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
355 356 357 358 359 
    "   $idata_kaldi/local_ESTER12/lm $idata_kaldi/local_ESTER12/lm/g2p $idata_kaldi/local_ESTER12/dict"
   ]
  },
  {
   "cell_type": "code",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
360 361 362 363 
   "execution_count": 15,
   "metadata": {
    "scrolled": true
   },
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
364 365 366 367 368 
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
369 370 371 
      "Checking data_last_ffmpeg/local_ESTER12/dict/silence_phones.txt ...\n",
      "--> reading data_last_ffmpeg/local_ESTER12/dict/silence_phones.txt\n",
      "--> data_last_ffmpeg/local_ESTER12/dict/silence_phones.txt is OK\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
372 
      "\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
373 374 375 
      "Checking data_last_ffmpeg/local_ESTER12/dict/optional_silence.txt ...\n",
      "--> reading data_last_ffmpeg/local_ESTER12/dict/optional_silence.txt\n",
      "--> data_last_ffmpeg/local_ESTER12/dict/optional_silence.txt is OK\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
376 
      "\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
377 378 379 
      "Checking data_last_ffmpeg/local_ESTER12/dict/nonsilence_phones.txt ...\n",
      "--> reading data_last_ffmpeg/local_ESTER12/dict/nonsilence_phones.txt\n",
      "--> data_last_ffmpeg/local_ESTER12/dict/nonsilence_phones.txt is OK\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
380 381 382 383 
      "\n",
      "Checking disjoint: silence_phones.txt, nonsilence_phones.txt\n",
      "--> disjoint property is OK.\n",
      "\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
384 385 386 
      "Checking data_last_ffmpeg/local_ESTER12/dict/lexicon.txt\n",
      "--> reading data_last_ffmpeg/local_ESTER12/dict/lexicon.txt\n",
      "--> data_last_ffmpeg/local_ESTER12/dict/lexicon.txt is OK\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
387 
      "\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
388 389 390 391 
      "Checking data_last_ffmpeg/local_ESTER12/dict/extra_questions.txt ...\n",
      "--> reading data_last_ffmpeg/local_ESTER12/dict/extra_questions.txt\n",
      "--> data_last_ffmpeg/local_ESTER12/dict/extra_questions.txt is OK\n",
      "--> SUCCESS [validating dictionary directory data_last_ffmpeg/local_ESTER12/dict]\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
392 
      "\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
393 394 
      "**Creating data_last_ffmpeg/local_ESTER12/dict/lexiconp.txt from data_last_ffmpeg/local_ESTER12/dict/lexicon.txt\n",
      "fstaddselfloops data_last_ffmpeg/lang_ESTER12/phones/wdisambig_phones.int data_last_ffmpeg/lang_ESTER12/phones/wdisambig_words.int \n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
395 
      "prepare_lang.sh: validating output directory\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
396 397 398 
      "utils/validate_lang.pl data_last_ffmpeg/lang_ESTER12\n",
      "Checking data_last_ffmpeg/lang_ESTER12/phones.txt ...\n",
      "--> data_last_ffmpeg/lang_ESTER12/phones.txt is OK\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
399 400 
      "\n",
      "Checking words.txt: #0 ...\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
401 
      "--> data_last_ffmpeg/lang_ESTER12/words.txt is OK\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
402 403 404 405 406 407 408 409 410 411 
      "\n",
      "Checking disjoint: silence.txt, nonsilence.txt, disambig.txt ...\n",
      "--> silence.txt and nonsilence.txt are disjoint\n",
      "--> silence.txt and disambig.txt are disjoint\n",
      "--> disambig.txt and nonsilence.txt are disjoint\n",
      "--> disjoint property is OK\n",
      "\n",
      "Checking sumation: silence.txt, nonsilence.txt, disambig.txt ...\n",
      "--> summation property is OK\n",
      "\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 
      "Checking data_last_ffmpeg/lang_ESTER12/phones/context_indep.{txt, int, csl} ...\n",
      "--> 20 entry/entries in data_last_ffmpeg/lang_ESTER12/phones/context_indep.txt\n",
      "--> data_last_ffmpeg/lang_ESTER12/phones/context_indep.int corresponds to data_last_ffmpeg/lang_ESTER12/phones/context_indep.txt\n",
      "--> data_last_ffmpeg/lang_ESTER12/phones/context_indep.csl corresponds to data_last_ffmpeg/lang_ESTER12/phones/context_indep.txt\n",
      "--> data_last_ffmpeg/lang_ESTER12/phones/context_indep.{txt, int, csl} are OK\n",
      "\n",
      "Checking data_last_ffmpeg/lang_ESTER12/phones/nonsilence.{txt, int, csl} ...\n",
      "--> 144 entry/entries in data_last_ffmpeg/lang_ESTER12/phones/nonsilence.txt\n",
      "--> data_last_ffmpeg/lang_ESTER12/phones/nonsilence.int corresponds to data_last_ffmpeg/lang_ESTER12/phones/nonsilence.txt\n",
      "--> data_last_ffmpeg/lang_ESTER12/phones/nonsilence.csl corresponds to data_last_ffmpeg/lang_ESTER12/phones/nonsilence.txt\n",
      "--> data_last_ffmpeg/lang_ESTER12/phones/nonsilence.{txt, int, csl} are OK\n",
      "\n",
      "Checking data_last_ffmpeg/lang_ESTER12/phones/silence.{txt, int, csl} ...\n",
      "--> 20 entry/entries in data_last_ffmpeg/lang_ESTER12/phones/silence.txt\n",
      "--> data_last_ffmpeg/lang_ESTER12/phones/silence.int corresponds to data_last_ffmpeg/lang_ESTER12/phones/silence.txt\n",
      "--> data_last_ffmpeg/lang_ESTER12/phones/silence.csl corresponds to data_last_ffmpeg/lang_ESTER12/phones/silence.txt\n",
      "--> data_last_ffmpeg/lang_ESTER12/phones/silence.{txt, int, csl} are OK\n",
      "\n",
      "Checking data_last_ffmpeg/lang_ESTER12/phones/optional_silence.{txt, int, csl} ...\n",
      "--> 1 entry/entries in data_last_ffmpeg/lang_ESTER12/phones/optional_silence.txt\n",
      "--> data_last_ffmpeg/lang_ESTER12/phones/optional_silence.int corresponds to data_last_ffmpeg/lang_ESTER12/phones/optional_silence.txt\n",
      "--> data_last_ffmpeg/lang_ESTER12/phones/optional_silence.csl corresponds to data_last_ffmpeg/lang_ESTER12/phones/optional_silence.txt\n",
      "--> data_last_ffmpeg/lang_ESTER12/phones/optional_silence.{txt, int, csl} are OK\n",
      "\n",
      "Checking data_last_ffmpeg/lang_ESTER12/phones/disambig.{txt, int, csl} ...\n",
      "--> 18 entry/entries in data_last_ffmpeg/lang_ESTER12/phones/disambig.txt\n",
      "--> data_last_ffmpeg/lang_ESTER12/phones/disambig.int corresponds to data_last_ffmpeg/lang_ESTER12/phones/disambig.txt\n",
      "--> data_last_ffmpeg/lang_ESTER12/phones/disambig.csl corresponds to data_last_ffmpeg/lang_ESTER12/phones/disambig.txt\n",
      "--> data_last_ffmpeg/lang_ESTER12/phones/disambig.{txt, int, csl} are OK\n",
      "\n",
      "Checking data_last_ffmpeg/lang_ESTER12/phones/roots.{txt, int} ...\n",
      "--> 40 entry/entries in data_last_ffmpeg/lang_ESTER12/phones/roots.txt\n",
      "--> data_last_ffmpeg/lang_ESTER12/phones/roots.int corresponds to data_last_ffmpeg/lang_ESTER12/phones/roots.txt\n",
      "--> data_last_ffmpeg/lang_ESTER12/phones/roots.{txt, int} are OK\n",
      "\n",
      "Checking data_last_ffmpeg/lang_ESTER12/phones/sets.{txt, int} ...\n",
      "--> 40 entry/entries in data_last_ffmpeg/lang_ESTER12/phones/sets.txt\n",
      "--> data_last_ffmpeg/lang_ESTER12/phones/sets.int corresponds to data_last_ffmpeg/lang_ESTER12/phones/sets.txt\n",
      "--> data_last_ffmpeg/lang_ESTER12/phones/sets.{txt, int} are OK\n",
      "\n",
      "Checking data_last_ffmpeg/lang_ESTER12/phones/extra_questions.{txt, int} ...\n",
      "--> 11 entry/entries in data_last_ffmpeg/lang_ESTER12/phones/extra_questions.txt\n",
      "--> data_last_ffmpeg/lang_ESTER12/phones/extra_questions.int corresponds to data_last_ffmpeg/lang_ESTER12/phones/extra_questions.txt\n",
      "--> data_last_ffmpeg/lang_ESTER12/phones/extra_questions.{txt, int} are OK\n",
      "\n",
      "Checking data_last_ffmpeg/lang_ESTER12/phones/word_boundary.{txt, int} ...\n",
      "--> 164 entry/entries in data_last_ffmpeg/lang_ESTER12/phones/word_boundary.txt\n",
      "--> data_last_ffmpeg/lang_ESTER12/phones/word_boundary.int corresponds to data_last_ffmpeg/lang_ESTER12/phones/word_boundary.txt\n",
      "--> data_last_ffmpeg/lang_ESTER12/phones/word_boundary.{txt, int} are OK\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
461 462 
      "\n",
      "Checking optional_silence.txt ...\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
463 464 
      "--> reading data_last_ffmpeg/lang_ESTER12/phones/optional_silence.txt\n",
      "--> data_last_ffmpeg/lang_ESTER12/phones/optional_silence.txt is OK\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
465 466 
      "\n",
      "Checking disambiguation symbols: #0 and #1\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
467 468 
      "--> data_last_ffmpeg/lang_ESTER12/phones/disambig.txt has \"#0\" and \"#1\"\n",
      "--> data_last_ffmpeg/lang_ESTER12/phones/disambig.txt is OK\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
469 470 471 472 
      "\n",
      "Checking topo ...\n",
      "\n",
      "Checking word_boundary.txt: silence.txt, nonsilence.txt, disambig.txt ...\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
473 474 475 
      "--> data_last_ffmpeg/lang_ESTER12/phones/word_boundary.txt doesn't include disambiguation symbols\n",
      "--> data_last_ffmpeg/lang_ESTER12/phones/word_boundary.txt is the union of nonsilence.txt and silence.txt\n",
      "--> data_last_ffmpeg/lang_ESTER12/phones/word_boundary.txt is OK\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
476 477 
      "\n",
      "Checking word-level disambiguation symbols...\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
478 
      "--> data_last_ffmpeg/lang_ESTER12/phones/wdisambig.txt exists (newer prepare_lang.sh)\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
479 
      "Checking word_boundary.int and disambig.int\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
480 
      "--> generating a 86 word sequence\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
481 482 
      "--> resulting phone sequence from L.fst corresponds to the word sequence\n",
      "--> L.fst is OK\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
483 
      "--> generating a 35 word sequence\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
484 485 486 
      "--> resulting phone sequence from L_disambig.fst corresponds to the word sequence\n",
      "--> L_disambig.fst is OK\n",
      "\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
487 488 489 490 
      "Checking data_last_ffmpeg/lang_ESTER12/oov.{txt, int} ...\n",
      "--> 1 entry/entries in data_last_ffmpeg/lang_ESTER12/oov.txt\n",
      "--> data_last_ffmpeg/lang_ESTER12/oov.int corresponds to data_last_ffmpeg/lang_ESTER12/oov.txt\n",
      "--> data_last_ffmpeg/lang_ESTER12/oov.{txt, int} are OK\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
491 
      "\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
492 493 494 
      "--> data_last_ffmpeg/lang_ESTER12/L.fst is olabel sorted\n",
      "--> data_last_ffmpeg/lang_ESTER12/L_disambig.fst is olabel sorted\n",
      "--> SUCCESS [validating lang directory data_last_ffmpeg/lang_ESTER12]\n"
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 
     ]
    }
   ],
   "source": [
    "#### Prepare Lang ==> L.fst Vocabulary's automate finite state\n",
    "utils/prepare_lang.sh $idata_kaldi/local_ESTER12/dict \\\n",
    "   \"<unk>\" $idata_kaldi/local_ESTER12/lang_tmp $idata_kaldi/lang_ESTER12"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln: impossible de créer le lien symbolique 'data-ESTER-V4/local_ESTER12/lm/lm_tgsphinx.arpa.gz': Le fichier existe\n",
      "ln: impossible de créer le lien symbolique 'data-ESTER-V4/local_ESTER12/lm/lm_french-small.arpa.gz': Le fichier existe\n",
      "arpa2fst --disambig-symbol=#0 --read-symbol-table=data-ESTER-V4/lang_ESTER12_test_fglarge/words.txt - data-ESTER-V4/lang_ESTER12_test_fglarge/G.fst \n",
      "LOG (arpa2fst:Read():arpa-file-parser.cc:90) Reading \\data\\ section.\n",
      "LOG (arpa2fst:Read():arpa-file-parser.cc:145) Reading \\1-grams: section.\n",
      "LOG (arpa2fst:Read():arpa-file-parser.cc:145) Reading \\2-grams: section.\n",
      "LOG (arpa2fst:Read():arpa-file-parser.cc:145) Reading \\3-grams: section.\n",
      "LOG (arpa2fst:Read():arpa-file-parser.cc:145) Reading \\4-grams: section.\n",
      "utils/validate_lang.pl data-ESTER-V4/lang_ESTER12_test_fglarge\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_fglarge/phones.txt ...\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/phones.txt is OK\n",
      "\n",
      "Checking words.txt: #0 ...\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/words.txt is OK\n",
      "\n",
      "Checking disjoint: silence.txt, nonsilence.txt, disambig.txt ...\n",
      "--> silence.txt and nonsilence.txt are disjoint\n",
      "--> silence.txt and disambig.txt are disjoint\n",
      "--> disambig.txt and nonsilence.txt are disjoint\n",
      "--> disjoint property is OK\n",
      "\n",
      "Checking sumation: silence.txt, nonsilence.txt, disambig.txt ...\n",
      "--> summation property is OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_fglarge/phones/context_indep.{txt, int, csl} ...\n",
      "--> 20 entry/entries in data-ESTER-V4/lang_ESTER12_test_fglarge/phones/context_indep.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/phones/context_indep.int corresponds to data-ESTER-V4/lang_ESTER12_test_fglarge/phones/context_indep.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/phones/context_indep.csl corresponds to data-ESTER-V4/lang_ESTER12_test_fglarge/phones/context_indep.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/phones/context_indep.{txt, int, csl} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_fglarge/phones/nonsilence.{txt, int, csl} ...\n",
      "--> 144 entry/entries in data-ESTER-V4/lang_ESTER12_test_fglarge/phones/nonsilence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/phones/nonsilence.int corresponds to data-ESTER-V4/lang_ESTER12_test_fglarge/phones/nonsilence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/phones/nonsilence.csl corresponds to data-ESTER-V4/lang_ESTER12_test_fglarge/phones/nonsilence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/phones/nonsilence.{txt, int, csl} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_fglarge/phones/silence.{txt, int, csl} ...\n",
      "--> 20 entry/entries in data-ESTER-V4/lang_ESTER12_test_fglarge/phones/silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/phones/silence.int corresponds to data-ESTER-V4/lang_ESTER12_test_fglarge/phones/silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/phones/silence.csl corresponds to data-ESTER-V4/lang_ESTER12_test_fglarge/phones/silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/phones/silence.{txt, int, csl} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_fglarge/phones/optional_silence.{txt, int, csl} ...\n",
      "--> 1 entry/entries in data-ESTER-V4/lang_ESTER12_test_fglarge/phones/optional_silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/phones/optional_silence.int corresponds to data-ESTER-V4/lang_ESTER12_test_fglarge/phones/optional_silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/phones/optional_silence.csl corresponds to data-ESTER-V4/lang_ESTER12_test_fglarge/phones/optional_silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/phones/optional_silence.{txt, int, csl} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_fglarge/phones/disambig.{txt, int, csl} ...\n",
      "--> 18 entry/entries in data-ESTER-V4/lang_ESTER12_test_fglarge/phones/disambig.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/phones/disambig.int corresponds to data-ESTER-V4/lang_ESTER12_test_fglarge/phones/disambig.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/phones/disambig.csl corresponds to data-ESTER-V4/lang_ESTER12_test_fglarge/phones/disambig.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/phones/disambig.{txt, int, csl} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_fglarge/phones/roots.{txt, int} ...\n",
      "--> 40 entry/entries in data-ESTER-V4/lang_ESTER12_test_fglarge/phones/roots.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/phones/roots.int corresponds to data-ESTER-V4/lang_ESTER12_test_fglarge/phones/roots.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/phones/roots.{txt, int} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_fglarge/phones/sets.{txt, int} ...\n",
      "--> 40 entry/entries in data-ESTER-V4/lang_ESTER12_test_fglarge/phones/sets.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/phones/sets.int corresponds to data-ESTER-V4/lang_ESTER12_test_fglarge/phones/sets.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/phones/sets.{txt, int} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_fglarge/phones/extra_questions.{txt, int} ...\n",
      "--> 11 entry/entries in data-ESTER-V4/lang_ESTER12_test_fglarge/phones/extra_questions.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/phones/extra_questions.int corresponds to data-ESTER-V4/lang_ESTER12_test_fglarge/phones/extra_questions.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/phones/extra_questions.{txt, int} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_fglarge/phones/word_boundary.{txt, int} ...\n",
      "--> 164 entry/entries in data-ESTER-V4/lang_ESTER12_test_fglarge/phones/word_boundary.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/phones/word_boundary.int corresponds to data-ESTER-V4/lang_ESTER12_test_fglarge/phones/word_boundary.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/phones/word_boundary.{txt, int} are OK\n",
      "\n",
      "Checking optional_silence.txt ...\n",
      "--> reading data-ESTER-V4/lang_ESTER12_test_fglarge/phones/optional_silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/phones/optional_silence.txt is OK\n",
      "\n",
      "Checking disambiguation symbols: #0 and #1\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/phones/disambig.txt has \"#0\" and \"#1\"\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/phones/disambig.txt is OK\n",
      "\n",
      "Checking topo ...\n",
      "\n",
      "Checking word_boundary.txt: silence.txt, nonsilence.txt, disambig.txt ...\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/phones/word_boundary.txt doesn't include disambiguation symbols\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/phones/word_boundary.txt is the union of nonsilence.txt and silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/phones/word_boundary.txt is OK\n",
      "\n",
      "Checking word-level disambiguation symbols...\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/phones/wdisambig.txt exists (newer prepare_lang.sh)\n",
      "Checking word_boundary.int and disambig.int\n",
      "--> generating a 55 word sequence\n",
      "--> resulting phone sequence from L.fst corresponds to the word sequence\n",
      "--> L.fst is OK\n",
      "--> generating a 56 word sequence\n",
      "--> resulting phone sequence from L_disambig.fst corresponds to the word sequence\n",
      "--> L_disambig.fst is OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_fglarge/oov.{txt, int} ...\n",
      "--> 1 entry/entries in data-ESTER-V4/lang_ESTER12_test_fglarge/oov.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/oov.int corresponds to data-ESTER-V4/lang_ESTER12_test_fglarge/oov.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/oov.{txt, int} are OK\n",
      "\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/L.fst is olabel sorted\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/L_disambig.fst is olabel sorted\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/G.fst is ilabel sorted\n",
      "--> data-ESTER-V4/lang_ESTER12_test_fglarge/G.fst has 721774 states\n",
      "--> utils/lang/check_g_properties.pl successfully validated data-ESTER-V4/lang_ESTER12_test_fglarge/G.fst\n",
      "--> utils/lang/check_g_properties.pl succeeded.\n",
      "--> SUCCESS [validating lang directory data-ESTER-V4/lang_ESTER12_test_fglarge]\n",
      "arpa2fst --disambig-symbol=#0 --read-symbol-table=data-ESTER-V4/lang_ESTER12_test_tglarge/words.txt - data-ESTER-V4/lang_ESTER12_test_tglarge/G.fst \n",
      "LOG (arpa2fst:Read():arpa-file-parser.cc:90) Reading \\data\\ section.\n",
      "LOG (arpa2fst:Read():arpa-file-parser.cc:145) Reading \\1-grams: section.\n",
      "LOG (arpa2fst:Read():arpa-file-parser.cc:145) Reading \\2-grams: section.\n",
      "LOG (arpa2fst:Read():arpa-file-parser.cc:145) Reading \\3-grams: section.\n",
      "utils/validate_lang.pl data-ESTER-V4/lang_ESTER12_test_tglarge\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Checking data-ESTER-V4/lang_ESTER12_test_tglarge/phones.txt ...\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/phones.txt is OK\n",
      "\n",
      "Checking words.txt: #0 ...\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/words.txt is OK\n",
      "\n",
      "Checking disjoint: silence.txt, nonsilence.txt, disambig.txt ...\n",
      "--> silence.txt and nonsilence.txt are disjoint\n",
      "--> silence.txt and disambig.txt are disjoint\n",
      "--> disambig.txt and nonsilence.txt are disjoint\n",
      "--> disjoint property is OK\n",
      "\n",
      "Checking sumation: silence.txt, nonsilence.txt, disambig.txt ...\n",
      "--> summation property is OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_tglarge/phones/context_indep.{txt, int, csl} ...\n",
      "--> 20 entry/entries in data-ESTER-V4/lang_ESTER12_test_tglarge/phones/context_indep.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/phones/context_indep.int corresponds to data-ESTER-V4/lang_ESTER12_test_tglarge/phones/context_indep.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/phones/context_indep.csl corresponds to data-ESTER-V4/lang_ESTER12_test_tglarge/phones/context_indep.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/phones/context_indep.{txt, int, csl} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_tglarge/phones/nonsilence.{txt, int, csl} ...\n",
      "--> 144 entry/entries in data-ESTER-V4/lang_ESTER12_test_tglarge/phones/nonsilence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/phones/nonsilence.int corresponds to data-ESTER-V4/lang_ESTER12_test_tglarge/phones/nonsilence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/phones/nonsilence.csl corresponds to data-ESTER-V4/lang_ESTER12_test_tglarge/phones/nonsilence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/phones/nonsilence.{txt, int, csl} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_tglarge/phones/silence.{txt, int, csl} ...\n",
      "--> 20 entry/entries in data-ESTER-V4/lang_ESTER12_test_tglarge/phones/silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/phones/silence.int corresponds to data-ESTER-V4/lang_ESTER12_test_tglarge/phones/silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/phones/silence.csl corresponds to data-ESTER-V4/lang_ESTER12_test_tglarge/phones/silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/phones/silence.{txt, int, csl} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_tglarge/phones/optional_silence.{txt, int, csl} ...\n",
      "--> 1 entry/entries in data-ESTER-V4/lang_ESTER12_test_tglarge/phones/optional_silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/phones/optional_silence.int corresponds to data-ESTER-V4/lang_ESTER12_test_tglarge/phones/optional_silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/phones/optional_silence.csl corresponds to data-ESTER-V4/lang_ESTER12_test_tglarge/phones/optional_silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/phones/optional_silence.{txt, int, csl} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_tglarge/phones/disambig.{txt, int, csl} ...\n",
      "--> 18 entry/entries in data-ESTER-V4/lang_ESTER12_test_tglarge/phones/disambig.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/phones/disambig.int corresponds to data-ESTER-V4/lang_ESTER12_test_tglarge/phones/disambig.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/phones/disambig.csl corresponds to data-ESTER-V4/lang_ESTER12_test_tglarge/phones/disambig.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/phones/disambig.{txt, int, csl} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_tglarge/phones/roots.{txt, int} ...\n",
      "--> 40 entry/entries in data-ESTER-V4/lang_ESTER12_test_tglarge/phones/roots.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/phones/roots.int corresponds to data-ESTER-V4/lang_ESTER12_test_tglarge/phones/roots.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/phones/roots.{txt, int} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_tglarge/phones/sets.{txt, int} ...\n",
      "--> 40 entry/entries in data-ESTER-V4/lang_ESTER12_test_tglarge/phones/sets.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/phones/sets.int corresponds to data-ESTER-V4/lang_ESTER12_test_tglarge/phones/sets.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/phones/sets.{txt, int} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_tglarge/phones/extra_questions.{txt, int} ...\n",
      "--> 11 entry/entries in data-ESTER-V4/lang_ESTER12_test_tglarge/phones/extra_questions.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/phones/extra_questions.int corresponds to data-ESTER-V4/lang_ESTER12_test_tglarge/phones/extra_questions.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/phones/extra_questions.{txt, int} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_tglarge/phones/word_boundary.{txt, int} ...\n",
      "--> 164 entry/entries in data-ESTER-V4/lang_ESTER12_test_tglarge/phones/word_boundary.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/phones/word_boundary.int corresponds to data-ESTER-V4/lang_ESTER12_test_tglarge/phones/word_boundary.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/phones/word_boundary.{txt, int} are OK\n",
      "\n",
      "Checking optional_silence.txt ...\n",
      "--> reading data-ESTER-V4/lang_ESTER12_test_tglarge/phones/optional_silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/phones/optional_silence.txt is OK\n",
      "\n",
      "Checking disambiguation symbols: #0 and #1\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/phones/disambig.txt has \"#0\" and \"#1\"\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/phones/disambig.txt is OK\n",
      "\n",
      "Checking topo ...\n",
      "\n",
      "Checking word_boundary.txt: silence.txt, nonsilence.txt, disambig.txt ...\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/phones/word_boundary.txt doesn't include disambiguation symbols\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/phones/word_boundary.txt is the union of nonsilence.txt and silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/phones/word_boundary.txt is OK\n",
      "\n",
      "Checking word-level disambiguation symbols...\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/phones/wdisambig.txt exists (newer prepare_lang.sh)\n",
      "Checking word_boundary.int and disambig.int\n",
      "--> generating a 66 word sequence\n",
      "--> resulting phone sequence from L.fst corresponds to the word sequence\n",
      "--> L.fst is OK\n",
      "--> generating a 92 word sequence\n",
      "--> resulting phone sequence from L_disambig.fst corresponds to the word sequence\n",
      "--> L_disambig.fst is OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_tglarge/oov.{txt, int} ...\n",
      "--> 1 entry/entries in data-ESTER-V4/lang_ESTER12_test_tglarge/oov.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/oov.int corresponds to data-ESTER-V4/lang_ESTER12_test_tglarge/oov.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/oov.{txt, int} are OK\n",
      "\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/L.fst is olabel sorted\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/L_disambig.fst is olabel sorted\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/G.fst is ilabel sorted\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tglarge/G.fst has 527277 states\n",
      "--> utils/lang/check_g_properties.pl successfully validated data-ESTER-V4/lang_ESTER12_test_tglarge/G.fst\n",
      "--> utils/lang/check_g_properties.pl succeeded.\n",
      "--> SUCCESS [validating lang directory data-ESTER-V4/lang_ESTER12_test_tglarge]\n",
      "arpa2fst --disambig-symbol=#0 --read-symbol-table=data-ESTER-V4/lang_ESTER12_test_tgsphinx/words.txt - data-ESTER-V4/lang_ESTER12_test_tgsphinx/G.fst \n",
      "LOG (arpa2fst:Read():arpa-file-parser.cc:90) Reading \\data\\ section.\n",
      "LOG (arpa2fst:Read():arpa-file-parser.cc:145) Reading \\1-grams: section.\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 8 [-5.2142\t-ce\t-0.2163] skipped: word '-ce' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 9 [-4.8723\t-ci\t-0.2944] skipped: word '-ci' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 10 [-4.2048\t-elle\t-0.3700] skipped: word '-elle' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 11 [-4.4310\t-elles\t-0.3539] skipped: word '-elles' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 12 [-4.9062\t-en\t-0.2800] skipped: word '-en' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 13 [-4.0463\t-il\t-0.4450] skipped: word '-il' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 14 [-4.1699\t-ils\t-0.4206] skipped: word '-ils' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 15 [-5.1270\t-je\t-0.2720] skipped: word '-je' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 16 [-4.9573\t-la\t-0.2149] skipped: word '-la' not in symbol table\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 17 [-3.6009\t-là\t-0.5456] skipped: word '-là' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 18 [-4.7289\t-le\t-0.2947] skipped: word '-le' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 19 [-5.0130\t-les\t-0.1999] skipped: word '-les' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 20 [-5.5490\t-lui\t-0.1649] skipped: word '-lui' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 21 [-4.7873\t-même\t-0.2425] skipped: word '-même' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 22 [-5.4825\t-mêmes\t-0.1766] skipped: word '-mêmes' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 23 [-5.0634\t-moi\t-0.3092] skipped: word '-moi' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 24 [-4.7140\t-nous\t-0.3254] skipped: word '-nous' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 25 [-4.3209\t-on\t-0.3717] skipped: word '-on' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 26 [-3.8887\t-t\t-1.4364] skipped: word '-t' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 27 [-4.3967\t-t-elle\t-0.2546] skipped: word '-t-elle' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 28 [-4.2829\t-t-il\t-0.2571] skipped: word '-t-il' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 29 [-5.3860\t-toi\t-0.1905] skipped: word '-toi' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 30 [-5.4007\t-tu\t-0.1668] skipped: word '-tu' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 31 [-4.6364\t-vous\t-0.3545] skipped: word '-vous' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 32 [-5.6027\t-y\t-0.1394] skipped: word '-y' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 38 [-5.8621\tà-côtés\t-0.1238] skipped: word 'à-côtés' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 39 [-5.7419\tà-coups\t-0.1460] skipped: word 'à-coups' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 40 [-5.7226\tà-peu-près\t-0.1351] skipped: word 'à-peu-près' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 41 [-5.8535\tà-propos\t-0.1315] skipped: word 'à-propos' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 42 [-5.3186\ta1\t-0.2572] skipped: word 'a1' not in symbol table\n",
      "LOG (arpa2fst:Read():arpa-file-parser.cc:145) Reading \\2-grams: section.\n",
      "LOG (arpa2fst:Read():arpa-file-parser.cc:145) Reading \\3-grams: section.\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:231) Of 9454889 parse warnings, 30 were reported. Run program with --max_warnings=-1 to see all warnings\n",
      "utils/validate_lang.pl data-ESTER-V4/lang_ESTER12_test_tgsphinx\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones.txt ...\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones.txt is OK\n",
      "\n",
      "Checking words.txt: #0 ...\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/words.txt is OK\n",
      "\n",
      "Checking disjoint: silence.txt, nonsilence.txt, disambig.txt ...\n",
      "--> silence.txt and nonsilence.txt are disjoint\n",
      "--> silence.txt and disambig.txt are disjoint\n",
      "--> disambig.txt and nonsilence.txt are disjoint\n",
      "--> disjoint property is OK\n",
      "\n",
      "Checking sumation: silence.txt, nonsilence.txt, disambig.txt ...\n",
      "--> summation property is OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/context_indep.{txt, int, csl} ...\n",
      "--> 20 entry/entries in data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/context_indep.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/context_indep.int corresponds to data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/context_indep.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/context_indep.csl corresponds to data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/context_indep.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/context_indep.{txt, int, csl} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/nonsilence.{txt, int, csl} ...\n",
      "--> 144 entry/entries in data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/nonsilence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/nonsilence.int corresponds to data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/nonsilence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/nonsilence.csl corresponds to data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/nonsilence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/nonsilence.{txt, int, csl} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/silence.{txt, int, csl} ...\n",
      "--> 20 entry/entries in data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/silence.int corresponds to data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/silence.csl corresponds to data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/silence.{txt, int, csl} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/optional_silence.{txt, int, csl} ...\n",
      "--> 1 entry/entries in data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/optional_silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/optional_silence.int corresponds to data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/optional_silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/optional_silence.csl corresponds to data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/optional_silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/optional_silence.{txt, int, csl} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/disambig.{txt, int, csl} ...\n",
      "--> 18 entry/entries in data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/disambig.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/disambig.int corresponds to data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/disambig.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/disambig.csl corresponds to data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/disambig.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/disambig.{txt, int, csl} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/roots.{txt, int} ...\n",
      "--> 40 entry/entries in data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/roots.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/roots.int corresponds to data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/roots.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/roots.{txt, int} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/sets.{txt, int} ...\n",
      "--> 40 entry/entries in data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/sets.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/sets.int corresponds to data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/sets.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/sets.{txt, int} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/extra_questions.{txt, int} ...\n",
      "--> 11 entry/entries in data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/extra_questions.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/extra_questions.int corresponds to data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/extra_questions.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/extra_questions.{txt, int} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/word_boundary.{txt, int} ...\n",
      "--> 164 entry/entries in data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/word_boundary.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/word_boundary.int corresponds to data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/word_boundary.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/word_boundary.{txt, int} are OK\n",
      "\n",
      "Checking optional_silence.txt ...\n",
      "--> reading data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/optional_silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/optional_silence.txt is OK\n",
      "\n",
      "Checking disambiguation symbols: #0 and #1\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/disambig.txt has \"#0\" and \"#1\"\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/disambig.txt is OK\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Checking topo ...\n",
      "\n",
      "Checking word_boundary.txt: silence.txt, nonsilence.txt, disambig.txt ...\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/word_boundary.txt doesn't include disambiguation symbols\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/word_boundary.txt is the union of nonsilence.txt and silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/word_boundary.txt is OK\n",
      "\n",
      "Checking word-level disambiguation symbols...\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/phones/wdisambig.txt exists (newer prepare_lang.sh)\n",
      "Checking word_boundary.int and disambig.int\n",
      "--> generating a 11 word sequence\n",
      "--> resulting phone sequence from L.fst corresponds to the word sequence\n",
      "--> L.fst is OK\n",
      "--> generating a 46 word sequence\n",
      "--> resulting phone sequence from L_disambig.fst corresponds to the word sequence\n",
      "--> L_disambig.fst is OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_tgsphinx/oov.{txt, int} ...\n",
      "--> 1 entry/entries in data-ESTER-V4/lang_ESTER12_test_tgsphinx/oov.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/oov.int corresponds to data-ESTER-V4/lang_ESTER12_test_tgsphinx/oov.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/oov.{txt, int} are OK\n",
      "\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/L.fst is olabel sorted\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/L_disambig.fst is olabel sorted\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/G.fst is ilabel sorted\n",
      "--> data-ESTER-V4/lang_ESTER12_test_tgsphinx/G.fst has 13156697 states\n",
      "--> utils/lang/check_g_properties.pl successfully validated data-ESTER-V4/lang_ESTER12_test_tgsphinx/G.fst\n",
      "--> utils/lang/check_g_properties.pl succeeded.\n",
      "--> SUCCESS [validating lang directory data-ESTER-V4/lang_ESTER12_test_tgsphinx]\n",
      "arpa2fst --disambig-symbol=#0 --read-symbol-table=data-ESTER-V4/lang_ESTER12_test_french-small/words.txt - data-ESTER-V4/lang_ESTER12_test_french-small/G.fst \n",
      "LOG (arpa2fst:Read():arpa-file-parser.cc:90) Reading \\data\\ section.\n",
      "LOG (arpa2fst:Read():arpa-file-parser.cc:145) Reading \\1-grams: section.\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 12 [-5.538307\taalto] skipped: word 'aalto' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 13 [-5.120474\taaron\t-0.02769727] skipped: word 'aaron' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 15 [-5.731214\tabachidze] skipped: word 'abachidze' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 16 [-5.487072\tabadie\t-0.02312817] skipped: word 'abadie' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 17 [-5.749587\tabagnale\t-0.03421882] skipped: word 'abagnale' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 18 [-5.439013\tabaissant\t-0.1023452] skipped: word 'abaissant' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 19 [-5.298798\tabaisse\t-0.1515498] skipped: word 'abaisse' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 23 [-5.383688\tabaissée\t-0.1352918] skipped: word 'abaissée' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 24 [-5.447463\tabaissés\t-0.1639998] skipped: word 'abaissés' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 26 [-5.323689\tabandonna\t-0.07547411] skipped: word 'abandonna' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 32 [-5.20608\tabandonnera\t-0.08380523] skipped: word 'abandonnera' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 42 [-5.217123\tabasourdi\t-0.1560311] skipped: word 'abasourdi' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 46 [-5.573611\tabat-jour\t-0.04034681] skipped: word 'abat-jour' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 47 [-5.525448\tabats\t-0.06064676] skipped: word 'abats' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 51 [-5.45354\tabattements\t-0.1281941] skipped: word 'abattements' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 61 [-5.026776\tabb\t-0.07580301] skipped: word 'abb' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 62 [-5.444225\tabbado\t-0.02391137] skipped: word 'abbado' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 64 [-5.533504\tabbatiale\t-0.05947962] skipped: word 'abbatiale' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 66 [-5.472561\tabbayes\t-0.06715237] skipped: word 'abbayes' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 67 [-5.64184\tabbesses\t-0.01320701] skipped: word 'abbesses' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 71 [-5.262563\tabbott\t-0.03037171] skipped: word 'abbott' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 89 [-5.243713\tabdelkrim\t-0.003165964] skipped: word 'abdelkrim' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 97 [-5.60272\tabderazak\t-0.07333852] skipped: word 'abderazak' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 102 [-5.590033\tabderrezak\t-0.08256684] skipped: word 'abderrezak' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 106 [-5.726902\tabdic] skipped: word 'abdic' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 107 [-5.558577\tabdication\t-0.05181389] skipped: word 'abdication' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 109 [-5.369082\tabdiquer\t-0.1094639] skipped: word 'abdiquer' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 110 [-5.468573\tabdiqué\t-0.07636042] skipped: word 'abdiqué' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 119 [-5.446184\tabe\t-0.003954563] skipped: word 'abe' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 120 [-5.70366\tabeau\t-0.02909388] skipped: word 'abeau' not in symbol table\n",
      "LOG (arpa2fst:Read():arpa-file-parser.cc:145) Reading \\2-grams: section.\n",
      "LOG (arpa2fst:Read():arpa-file-parser.cc:145) Reading \\3-grams: section.\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:231) Of 292476 parse warnings, 30 were reported. Run program with --max_warnings=-1 to see all warnings\n",
      "utils/validate_lang.pl data-ESTER-V4/lang_ESTER12_test_french-small\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_french-small/phones.txt ...\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/phones.txt is OK\n",
      "\n",
      "Checking words.txt: #0 ...\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/words.txt is OK\n",
      "\n",
      "Checking disjoint: silence.txt, nonsilence.txt, disambig.txt ...\n",
      "--> silence.txt and nonsilence.txt are disjoint\n",
      "--> silence.txt and disambig.txt are disjoint\n",
      "--> disambig.txt and nonsilence.txt are disjoint\n",
      "--> disjoint property is OK\n",
      "\n",
      "Checking sumation: silence.txt, nonsilence.txt, disambig.txt ...\n",
      "--> summation property is OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_french-small/phones/context_indep.{txt, int, csl} ...\n",
      "--> 20 entry/entries in data-ESTER-V4/lang_ESTER12_test_french-small/phones/context_indep.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/phones/context_indep.int corresponds to data-ESTER-V4/lang_ESTER12_test_french-small/phones/context_indep.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/phones/context_indep.csl corresponds to data-ESTER-V4/lang_ESTER12_test_french-small/phones/context_indep.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/phones/context_indep.{txt, int, csl} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_french-small/phones/nonsilence.{txt, int, csl} ...\n",
      "--> 144 entry/entries in data-ESTER-V4/lang_ESTER12_test_french-small/phones/nonsilence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/phones/nonsilence.int corresponds to data-ESTER-V4/lang_ESTER12_test_french-small/phones/nonsilence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/phones/nonsilence.csl corresponds to data-ESTER-V4/lang_ESTER12_test_french-small/phones/nonsilence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/phones/nonsilence.{txt, int, csl} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_french-small/phones/silence.{txt, int, csl} ...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--> 20 entry/entries in data-ESTER-V4/lang_ESTER12_test_french-small/phones/silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/phones/silence.int corresponds to data-ESTER-V4/lang_ESTER12_test_french-small/phones/silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/phones/silence.csl corresponds to data-ESTER-V4/lang_ESTER12_test_french-small/phones/silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/phones/silence.{txt, int, csl} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_french-small/phones/optional_silence.{txt, int, csl} ...\n",
      "--> 1 entry/entries in data-ESTER-V4/lang_ESTER12_test_french-small/phones/optional_silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/phones/optional_silence.int corresponds to data-ESTER-V4/lang_ESTER12_test_french-small/phones/optional_silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/phones/optional_silence.csl corresponds to data-ESTER-V4/lang_ESTER12_test_french-small/phones/optional_silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/phones/optional_silence.{txt, int, csl} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_french-small/phones/disambig.{txt, int, csl} ...\n",
      "--> 18 entry/entries in data-ESTER-V4/lang_ESTER12_test_french-small/phones/disambig.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/phones/disambig.int corresponds to data-ESTER-V4/lang_ESTER12_test_french-small/phones/disambig.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/phones/disambig.csl corresponds to data-ESTER-V4/lang_ESTER12_test_french-small/phones/disambig.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/phones/disambig.{txt, int, csl} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_french-small/phones/roots.{txt, int} ...\n",
      "--> 40 entry/entries in data-ESTER-V4/lang_ESTER12_test_french-small/phones/roots.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/phones/roots.int corresponds to data-ESTER-V4/lang_ESTER12_test_french-small/phones/roots.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/phones/roots.{txt, int} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_french-small/phones/sets.{txt, int} ...\n",
      "--> 40 entry/entries in data-ESTER-V4/lang_ESTER12_test_french-small/phones/sets.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/phones/sets.int corresponds to data-ESTER-V4/lang_ESTER12_test_french-small/phones/sets.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/phones/sets.{txt, int} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_french-small/phones/extra_questions.{txt, int} ...\n",
      "--> 11 entry/entries in data-ESTER-V4/lang_ESTER12_test_french-small/phones/extra_questions.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/phones/extra_questions.int corresponds to data-ESTER-V4/lang_ESTER12_test_french-small/phones/extra_questions.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/phones/extra_questions.{txt, int} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_french-small/phones/word_boundary.{txt, int} ...\n",
      "--> 164 entry/entries in data-ESTER-V4/lang_ESTER12_test_french-small/phones/word_boundary.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/phones/word_boundary.int corresponds to data-ESTER-V4/lang_ESTER12_test_french-small/phones/word_boundary.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/phones/word_boundary.{txt, int} are OK\n",
      "\n",
      "Checking optional_silence.txt ...\n",
      "--> reading data-ESTER-V4/lang_ESTER12_test_french-small/phones/optional_silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/phones/optional_silence.txt is OK\n",
      "\n",
      "Checking disambiguation symbols: #0 and #1\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/phones/disambig.txt has \"#0\" and \"#1\"\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/phones/disambig.txt is OK\n",
      "\n",
      "Checking topo ...\n",
      "\n",
      "Checking word_boundary.txt: silence.txt, nonsilence.txt, disambig.txt ...\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/phones/word_boundary.txt doesn't include disambiguation symbols\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/phones/word_boundary.txt is the union of nonsilence.txt and silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/phones/word_boundary.txt is OK\n",
      "\n",
      "Checking word-level disambiguation symbols...\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/phones/wdisambig.txt exists (newer prepare_lang.sh)\n",
      "Checking word_boundary.int and disambig.int\n",
      "--> generating a 25 word sequence\n",
      "--> resulting phone sequence from L.fst corresponds to the word sequence\n",
      "--> L.fst is OK\n",
      "--> generating a 43 word sequence\n",
      "--> resulting phone sequence from L_disambig.fst corresponds to the word sequence\n",
      "--> L_disambig.fst is OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_french-small/oov.{txt, int} ...\n",
      "--> 1 entry/entries in data-ESTER-V4/lang_ESTER12_test_french-small/oov.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/oov.int corresponds to data-ESTER-V4/lang_ESTER12_test_french-small/oov.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/oov.{txt, int} are OK\n",
      "\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/L.fst is olabel sorted\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/L_disambig.fst is olabel sorted\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/G.fst is ilabel sorted\n",
      "--> data-ESTER-V4/lang_ESTER12_test_french-small/G.fst has 878846 states\n",
      "--> utils/lang/check_g_properties.pl successfully validated data-ESTER-V4/lang_ESTER12_test_french-small/G.fst\n",
      "--> utils/lang/check_g_properties.pl succeeded.\n",
      "--> SUCCESS [validating lang directory data-ESTER-V4/lang_ESTER12_test_french-small]\n",
      "arpa2fst --disambig-symbol=#0 --read-symbol-table=data-ESTER-V4/lang_ESTER12_test_mixed/words.txt - data-ESTER-V4/lang_ESTER12_test_mixed/G.fst \n",
      "LOG (arpa2fst:Read():arpa-file-parser.cc:90) Reading \\data\\ section.\n",
      "LOG (arpa2fst:Read():arpa-file-parser.cc:145) Reading \\1-grams: section.\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 8 [-5.437585\t-ce\t-0.1377075] skipped: word '-ce' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 9 [-5.095685\t-ci\t-0.1822014] skipped: word '-ci' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 10 [-4.428185\t-elle\t-0.2227153] skipped: word '-elle' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 11 [-4.654386\t-elles\t-0.2134213] skipped: word '-elles' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 12 [-5.129585\t-en\t-0.1693132] skipped: word '-en' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 13 [-4.269685\t-il\t-0.2610333] skipped: word '-il' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 14 [-4.393285\t-ils\t-0.2481304] skipped: word '-ils' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 15 [-5.350385\t-je\t-0.170345] skipped: word '-je' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 16 [-5.180686\t-la\t-0.1257521] skipped: word '-la' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 17 [-4.952285\t-le\t-0.1823167] skipped: word '-le' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 18 [-5.236385\t-les\t-0.1255084] skipped: word '-les' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 19 [-5.772385\t-lui\t-0.1013363] skipped: word '-lui' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 20 [-3.824285\t-là\t-0.3025716] skipped: word '-là' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 21 [-5.286785\t-moi\t-0.1897822] skipped: word '-moi' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 22 [-5.010685\t-même\t-0.1537597] skipped: word '-même' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 23 [-5.705885\t-mêmes\t-0.1144569] skipped: word '-mêmes' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 24 [-4.937386\t-nous\t-0.1997344] skipped: word '-nous' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 25 [-4.544285\t-on\t-0.222972] skipped: word '-on' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 26 [-4.112085\t-t\t-0.3957374] skipped: word '-t' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 27 [-4.620085\t-t-elle\t-0.1606998] skipped: word '-t-elle' not in symbol table\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 28 [-4.506285\t-t-il\t-0.1620795] skipped: word '-t-il' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 29 [-5.609385\t-toi\t-0.1206716] skipped: word '-toi' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 30 [-5.624085\t-tu\t-0.1071467] skipped: word '-tu' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 31 [-4.859786\t-vous\t-0.21523] skipped: word '-vous' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 32 [-5.826086\t-y\t-0.08851968] skipped: word '-y' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 42 [-5.541986\ta1\t-0.1538226] skipped: word 'a1' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 43 [-5.912485\ta10\t-0.1032059] skipped: word 'a10' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 44 [-5.919685\ta104\t-0.07961074] skipped: word 'a104' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 45 [-6.224785\ta11\t-0.06672045] skipped: word 'a11' not in symbol table\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:192) line 46 [-6.026186\ta13\t-0.1179137] skipped: word 'a13' not in symbol table\n",
      "LOG (arpa2fst:Read():arpa-file-parser.cc:145) Reading \\2-grams: section.\n",
      "LOG (arpa2fst:Read():arpa-file-parser.cc:145) Reading \\3-grams: section.\n",
      "WARNING (arpa2fst:Read():arpa-file-parser.cc:231) Of 9454889 parse warnings, 30 were reported. Run program with --max_warnings=-1 to see all warnings\n",
      "utils/validate_lang.pl data-ESTER-V4/lang_ESTER12_test_mixed\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_mixed/phones.txt ...\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/phones.txt is OK\n",
      "\n",
      "Checking words.txt: #0 ...\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/words.txt is OK\n",
      "\n",
      "Checking disjoint: silence.txt, nonsilence.txt, disambig.txt ...\n",
      "--> silence.txt and nonsilence.txt are disjoint\n",
      "--> silence.txt and disambig.txt are disjoint\n",
      "--> disambig.txt and nonsilence.txt are disjoint\n",
      "--> disjoint property is OK\n",
      "\n",
      "Checking sumation: silence.txt, nonsilence.txt, disambig.txt ...\n",
      "--> summation property is OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_mixed/phones/context_indep.{txt, int, csl} ...\n",
      "--> 20 entry/entries in data-ESTER-V4/lang_ESTER12_test_mixed/phones/context_indep.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/phones/context_indep.int corresponds to data-ESTER-V4/lang_ESTER12_test_mixed/phones/context_indep.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/phones/context_indep.csl corresponds to data-ESTER-V4/lang_ESTER12_test_mixed/phones/context_indep.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/phones/context_indep.{txt, int, csl} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_mixed/phones/nonsilence.{txt, int, csl} ...\n",
      "--> 144 entry/entries in data-ESTER-V4/lang_ESTER12_test_mixed/phones/nonsilence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/phones/nonsilence.int corresponds to data-ESTER-V4/lang_ESTER12_test_mixed/phones/nonsilence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/phones/nonsilence.csl corresponds to data-ESTER-V4/lang_ESTER12_test_mixed/phones/nonsilence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/phones/nonsilence.{txt, int, csl} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_mixed/phones/silence.{txt, int, csl} ...\n",
      "--> 20 entry/entries in data-ESTER-V4/lang_ESTER12_test_mixed/phones/silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/phones/silence.int corresponds to data-ESTER-V4/lang_ESTER12_test_mixed/phones/silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/phones/silence.csl corresponds to data-ESTER-V4/lang_ESTER12_test_mixed/phones/silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/phones/silence.{txt, int, csl} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_mixed/phones/optional_silence.{txt, int, csl} ...\n",
      "--> 1 entry/entries in data-ESTER-V4/lang_ESTER12_test_mixed/phones/optional_silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/phones/optional_silence.int corresponds to data-ESTER-V4/lang_ESTER12_test_mixed/phones/optional_silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/phones/optional_silence.csl corresponds to data-ESTER-V4/lang_ESTER12_test_mixed/phones/optional_silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/phones/optional_silence.{txt, int, csl} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_mixed/phones/disambig.{txt, int, csl} ...\n",
      "--> 18 entry/entries in data-ESTER-V4/lang_ESTER12_test_mixed/phones/disambig.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/phones/disambig.int corresponds to data-ESTER-V4/lang_ESTER12_test_mixed/phones/disambig.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/phones/disambig.csl corresponds to data-ESTER-V4/lang_ESTER12_test_mixed/phones/disambig.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/phones/disambig.{txt, int, csl} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_mixed/phones/roots.{txt, int} ...\n",
      "--> 40 entry/entries in data-ESTER-V4/lang_ESTER12_test_mixed/phones/roots.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/phones/roots.int corresponds to data-ESTER-V4/lang_ESTER12_test_mixed/phones/roots.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/phones/roots.{txt, int} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_mixed/phones/sets.{txt, int} ...\n",
      "--> 40 entry/entries in data-ESTER-V4/lang_ESTER12_test_mixed/phones/sets.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/phones/sets.int corresponds to data-ESTER-V4/lang_ESTER12_test_mixed/phones/sets.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/phones/sets.{txt, int} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_mixed/phones/extra_questions.{txt, int} ...\n",
      "--> 11 entry/entries in data-ESTER-V4/lang_ESTER12_test_mixed/phones/extra_questions.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/phones/extra_questions.int corresponds to data-ESTER-V4/lang_ESTER12_test_mixed/phones/extra_questions.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/phones/extra_questions.{txt, int} are OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_mixed/phones/word_boundary.{txt, int} ...\n",
      "--> 164 entry/entries in data-ESTER-V4/lang_ESTER12_test_mixed/phones/word_boundary.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/phones/word_boundary.int corresponds to data-ESTER-V4/lang_ESTER12_test_mixed/phones/word_boundary.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/phones/word_boundary.{txt, int} are OK\n",
      "\n",
      "Checking optional_silence.txt ...\n",
      "--> reading data-ESTER-V4/lang_ESTER12_test_mixed/phones/optional_silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/phones/optional_silence.txt is OK\n",
      "\n",
      "Checking disambiguation symbols: #0 and #1\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/phones/disambig.txt has \"#0\" and \"#1\"\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/phones/disambig.txt is OK\n",
      "\n",
      "Checking topo ...\n",
      "\n",
      "Checking word_boundary.txt: silence.txt, nonsilence.txt, disambig.txt ...\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/phones/word_boundary.txt doesn't include disambiguation symbols\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/phones/word_boundary.txt is the union of nonsilence.txt and silence.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/phones/word_boundary.txt is OK\n",
      "\n",
      "Checking word-level disambiguation symbols...\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/phones/wdisambig.txt exists (newer prepare_lang.sh)\n",
      "Checking word_boundary.int and disambig.int\n",
      "--> generating a 60 word sequence\n",
      "--> resulting phone sequence from L.fst corresponds to the word sequence\n",
      "--> L.fst is OK\n",
      "--> generating a 72 word sequence\n",
      "--> resulting phone sequence from L_disambig.fst corresponds to the word sequence\n",
      "--> L_disambig.fst is OK\n",
      "\n",
      "Checking data-ESTER-V4/lang_ESTER12_test_mixed/oov.{txt, int} ...\n",
      "--> 1 entry/entries in data-ESTER-V4/lang_ESTER12_test_mixed/oov.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/oov.int corresponds to data-ESTER-V4/lang_ESTER12_test_mixed/oov.txt\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/oov.{txt, int} are OK\n",
      "\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/L.fst is olabel sorted\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/L_disambig.fst is olabel sorted\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/G.fst is ilabel sorted\n",
      "--> data-ESTER-V4/lang_ESTER12_test_mixed/G.fst has 13239100 states\n",
      "--> utils/lang/check_g_properties.pl successfully validated data-ESTER-V4/lang_ESTER12_test_mixed/G.fst\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--> utils/lang/check_g_properties.pl succeeded.\n",
      "--> SUCCESS [validating lang directory data-ESTER-V4/lang_ESTER12_test_mixed]\n",
      "Succeeded in formatting data.\n"
     ]
    }
   ],
   "source": [
    "#### Prepare Contextual automate finite state using LM's ===> Build G.fst\n",
    "#### Copy the largest LMS built from LIUM and generate G.fst :\n",
    "# Large LM\n",
    "ln -s $LM_LIUM_LARGE $idata_kaldi/local_ESTER12/lm/\n",
    "# Pruned LM\n",
    "ln -s $LM_LIUM_SMALL $idata_kaldi/local_ESTER12/lm/\n",
    "# Mixed\n",
    "dir_eval_lm=$exp_kaldi/eval_LM_ESTER12\n",
    "#mv $dir_eval_lm/mixed.gz $dir_eval_lm/lm_mixed.arpa.gz\n",
    "mv $dir_eval_lm/mixed.gz $idata_kaldi/local_ESTER12/lm/lm_mixed.arpa.gz\n",
    "local/format_lms.sh --src-dir $idata_kaldi/lang_ESTER12 $idata_kaldi/local_ESTER12/lm"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Recette"
   ]
  },
  {
   "cell_type": "code",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
1212 
   "execution_count": 16,
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
1213 1214 1215 1216 1217 1218 1219 1220 
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
1221 
      "steps/train_mono.sh --boost-silence 1.25 --nj 32 --cmd run.pl --mem 64G data_last_ffmpeg/ESTER1_2 data_last_ffmpeg/lang_ESTER12 exp_last_ffmpeg/monoAllESTER12\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
1222 1223 
      "filter_scps.pl: warning: some input lines were output to multiple files\n",
      "steps/train_mono.sh: Initializing monophone system.\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 
      "feat-to-dim 'ark,s,cs:apply-cmvn --utt2spk=ark:data_last_ffmpeg/ESTER1_2/split32/1/utt2spk scp:data_last_ffmpeg/ESTER1_2/split32/1/cmvn.scp scp:data_last_ffmpeg/ESTER1_2/split32/1/feats.scp ark:- | add-deltas ark:- ark:- |' - \n",
      "add-deltas ark:- ark:- \n",
      "apply-cmvn --utt2spk=ark:data_last_ffmpeg/ESTER1_2/split32/1/utt2spk scp:data_last_ffmpeg/ESTER1_2/split32/1/cmvn.scp scp:data_last_ffmpeg/ESTER1_2/split32/1/feats.scp ark:- \n",
      "WARNING (apply-cmvn:Open():util/kaldi-table-inl.h:106) Failed to open script file data_last_ffmpeg/ESTER1_2/split32/1/feats.scp\n",
      "ERROR (apply-cmvn:SequentialTableReader():util/kaldi-table-inl.h:876) Error constructing TableReader: rspecifier is scp:data_last_ffmpeg/ESTER1_2/split32/1/feats.scp\n",
      "\n",
      "[ Stack-Trace: ]\n",
      "apply-cmvn() [0x510766]\n",
      "kaldi::MessageLogger::HandleMessage(kaldi::LogMessageEnvelope const&, char const*)\n",
      "kaldi::MessageLogger::~MessageLogger()\n",
      "kaldi::SequentialTableReader<kaldi::KaldiObjectHolder<kaldi::Matrix<float> > >::SequentialTableReader(std::string const&)\n",
      "main\n",
      "__libc_start_main\n",
      "apply-cmvn() [0x466439]\n",
      "\n",
      "ERROR (feat-to-dim:main():feat-to-dim.cc:58) Could not read any features (empty archive?)\n",
      "\n",
      "[ Stack-Trace: ]\n",
      "feat-to-dim() [0x4ea3c2]\n",
      "kaldi::MessageLogger::HandleMessage(kaldi::LogMessageEnvelope const&, char const*)\n",
      "kaldi::MessageLogger::~MessageLogger()\n",
      "main\n",
      "__libc_start_main\n",
      "feat-to-dim() [0x451dc9]\n",
      "\n",
      "error getting feature dimension\n",
      "steps/train_deltas.sh --boost-silence 1.25 --cmd run.pl --mem 64G 10000 150000 data_last_ffmpeg/ESTER1_2 data_last_ffmpeg/lang_ESTER12 exp_last_ffmpeg/monoAllESTER12 exp_last_ffmpeg/monoAllESTER12/tri1_10K_100K_ESTER_All\n",
      "train_deltas.sh: no such file exp_last_ffmpeg/monoAllESTER12/final.mdl\n",
      "steps/train_deltas.sh --boost-silence 1.25 --cmd run.pl --mem 64G 15000 250000 data_last_ffmpeg/ESTER1_2 data_last_ffmpeg/lang_ESTER12 exp_last_ffmpeg/monoAllESTER12/tri1_10K_100K_ESTER_All exp_last_ffmpeg/monoAllESTER12/tri1_10K_100K_ESTER_All/tri1_15K_200K_ESTER_All\n",
      "train_deltas.sh: no such file exp_last_ffmpeg/monoAllESTER12/tri1_10K_100K_ESTER_All/final.mdl\n",
      "steps/train_lda_mllt.sh --cmd run.pl --mem 64G --splice-opts --left-context=8 --right-context=8 20000 350000 data_last_ffmpeg/ESTER1_2 data_last_ffmpeg/lang_ESTER12 exp_last_ffmpeg/monoAllESTER12/tri1_10K_100K_ESTER_All/tri1_15K_200K_ESTER_All exp_last_ffmpeg/monoAllESTER12/tri1_10K_100K_ESTER_All/tri1_15K_200K_ESTER_All/tri2_20K_300K_ESTER_All\n",
      "train_lda_mllt.sh: no such file exp_last_ffmpeg/monoAllESTER12/tri1_10K_100K_ESTER_All/tri1_15K_200K_ESTER_All/final.mdl\n"
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
1256 
     ]
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
1257 1258 1259 1260 1261 1262 
    },
    {
     "ename": "",
     "evalue": "1",
     "output_type": "error",
     "traceback": []
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
1263 1264 1265 1266 1267 1268 1269 1270 1271 
    }
   ],
   "source": [
    "# Execution rapide des modèles\n",
    "# avec french-small\n",
    "# Monophone OK\n",
    "#monoALL Phase 1 & 2\n",
    "exp_mono=$exp_kaldi/monoAllESTER12\n",
    " steps/train_mono.sh --boost-silence 1.25 --nj 32 --cmd \"$train_cmd\" \\\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
1272 
    "   $idata_kaldi/ESTER1_2 $idata_kaldi/lang_ESTER12 $exp_mono\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
1273 1274 1275 1276 1277 
    "# Go triphone\n",
    "num_leaves=10000\n",
    "tot_gauss=150000\n",
    "tri1_step1=$exp_mono/tri1_10K_100K_ESTER_All\n",
    "steps/train_deltas.sh --boost-silence 1.25 --cmd \"$train_cmd\" \\\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
1278 
    "     $num_leaves $tot_gauss $idata_kaldi/ESTER1_2 $idata_kaldi/lang_ESTER12 $exp_mono \\\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
1279 1280 1281 1282 1283 1284 
    "     $tri1_step1\n",
    "# seconde step triphone\n",
    "num_leaves=15000\n",
    "tot_gauss=250000\n",
    "tri1_step2=$tri1_step1/tri1_15K_200K_ESTER_All\n",
    "steps/train_deltas.sh --boost-silence 1.25 --cmd \"$train_cmd\" \\\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
1285 
    "     $num_leaves $tot_gauss $idata_kaldi/ESTER1_2 $idata_kaldi/lang_ESTER12 $tri1_step1 \\\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
1286 1287 1288 1289 1290 1291 1292 1293 1294 
    "     $tri1_step2\n",
    "context_dependence=\"--left-context=8 --right-context=8\"\n",
    "num_leaves=20000\n",
    "tot_gauss=350000\n",
    "tri2_step1=$tri1_step2/tri2_20K_300K_ESTER_All\n",
    "# Best Triphone model\n",
    "# Train lda_mllt based-on best models\n",
    "steps/train_lda_mllt.sh --cmd \"$train_cmd\" \\\n",
    "    --splice-opts \"$context_dependence\" $num_leaves $tot_gauss \\\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
1295 
    "    $idata_kaldi/ESTER1_2 \\\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 
    "    $idata_kaldi/lang_ESTER12 \\\n",
    "    $tri1_step2 \\\n",
    "    $tri2_step1\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# SAT training"
   ]
  },
  {
   "cell_type": "code",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
1311 
   "execution_count": null,
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
1312 
   "metadata": {
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
1313 
    "collapsed": true,
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
1314 1315 
    "scrolled": true
   },
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
1316 
   "outputs": [],
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
1317 1318 1319 
   "source": [
    "#exp_lda=exp-ESTER-V2-noise/monoAll_1_2/tri1_20000K_100000K_ESTER_All/tri1_40000K_100000K_ESTER_All/tri1_LDA_50K_300K\n",
    "# Restart\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
1320 
    "#exp_kaldi=exp-ESTER-V4\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
1321 1322 1323 1324 1325 1326 
    "exp_mono=$exp_kaldi/monoAllESTER12\n",
    "tri1_step1=$exp_mono/tri1_10K_100K_ESTER_All\n",
    "tri1_step2=$tri1_step1/tri1_15K_200K_ESTER_All\n",
    "tri2_step1=$tri1_step2/tri2_20K_300K_ESTER_All\n",
    "tri3_ali=$tri2_step1/tri3_align_fmllr\n",
    "steps/align_fmllr.sh --nj 32 --cmd \"$train_cmd\" \\\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
1327 
    "   $idata_kaldi/ESTER1_2 $idata_kaldi/lang_ESTER12 $tri2_step1 $tri3_ali\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
1328 1329 
    "tri3_step1=$tri2_step1/tri3_SAT_50K_400K\n",
    " steps/train_sat.sh  --cmd \"$train_cmd\" 50000 450000 \\\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
1330 
    "   $idata_kaldi/ESTER1_2 $idata_kaldi/lang_ESTER12 $tri3_ali $tri3_step1\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
1331 1332 
    "tri3_step2=$tri3_step1/tri3_SAT_70K_500K\n",
    "steps/train_sat.sh  --cmd \"$train_cmd\" 70000 550000 \\\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
1333 
    " $idata_kaldi/ESTER1_2 $idata_kaldi/lang_ESTER12 $tri3_step1 \\\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
1334 1335 1336 1337 1338 
    " $tri3_step2"
   ]
  },
  {
   "cell_type": "code",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
1339 1340 1341 1342 1343 
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
1344 1345 
   "source": [
    "steps/get_prons.sh --cmd \"$train_cmd\" \\\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
1346 
    "   $idata_kaldi/ESTER1_2 $idata_kaldi/lang_ESTER12 $tri3_step2\n",
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
1347 1348 1349 1350 1351 1352 1353 
    " utils/dict_dir_add_pronprobs.sh --max-normalize true \\\n",
    "   $idata_kaldi/local_ESTER12/dict \\\n",
    "   $tri3_step2/pron_counts_nowb.txt $tri3_step2/sil_counts_nowb.txt \\\n",
    "   $tri3_step2/pron_bigram_counts_nowb.txt $idata_kaldi/local_ESTER12/dict_new_sat\n",
    "\n",
    " utils/prepare_lang.sh $idata_kaldi/local_ESTER12/dict_new_sat \\\n",
    "   \"<unk>\" $idata_kaldi/local_ESTER12/lang_tmp_new_sat $idata_kaldi/lang_ESTER12_new_sat\n",
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
1354 
    " #local/format_lms.sh --src-dir $idata_kaldi/lang_ESTER12_new_sat $idata_kaldi/local_ESTER12/lm"
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
1355 1356 1357 1358 1359 1360 
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
1361 
    "collapsed": true,
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
1362 1363 
    "scrolled": true
   },
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
1364 
   "outputs": [],
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 
   "source": [
    "exp_mono=$exp_kaldi/monoAllESTER12\n",
    "tri1_step1=$exp_mono/tri1_10K_100K_ESTER_All\n",
    "tri1_step2=$tri1_step1/tri1_15K_200K_ESTER_All\n",
    "tri2_step1=$tri1_step2/tri2_20K_300K_ESTER_All\n",
    "tri3_ali=$tri2_step1/tri1_align_fmllr\n",
    "tri3_step1=$tri2_step1/tri3_SAT_50K_400K\n",
    "tri3_step2=$tri3_step1/tri3_SAT_70K_500K\n",
    "#exp_sat=$exp_lda/tri1_align_fmllr/tri1_SAT_50K_400K/tri1_SAT_70K_500K\n",
    "for lg in tglarge fglarge; do\n",
    "#utils/mkgraph.sh $idata_kaldi/lang_ESTER12_new_sat_test_$lg \\\n",
    "#     $tri3_step2 $tri3_step2/graph_lang_ESTER12_sat_$lg\n",
    "steps/decode_fmllr.sh --nj 32 --cmd \"$decode_cmd\" \\\n",
    "       $tri3_step2/graph_lang_ESTER12_sat_$lg $idata_kaldi/DATA \\\n",
    "       $tri3_step2/decode_EVAL2005_sat_$lg\n",
    "done"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "exp_mono=$exp_kaldi/monoAllESTER12\n",
    "tri1_step1=$exp_mono/tri1_10K_100K_ESTER_All\n",
    "tri1_step2=$tri1_step1/tri1_15K_200K_ESTER_All\n",
    "tri2_step1=$tri1_step2/tri2_20K_300K_ESTER_All\n",
    "tri3_ali=$tri2_step1/tri1_align_fmllr\n",
    "tri3_step1=$tri2_step1/tri3_SAT_50K_400K\n",
    "tri3_step2=$tri3_step1/tri3_SAT_70K_500K\n",
    "#exp_sat=$exp_lda/tri1_align_fmllr/tri1_SAT_50K_400K/tri1_SAT_70K_500K\n",
    "for lg in tgsphinx mixed; do\n",
    "#utils/mkgraph.sh $idata_kaldi/lang_ESTER12_new_sat_test_$lg \\\n",
    "#     $tri3_step2 $tri3_step2/graph_lang_ESTER12_sat_$lg\n",
    "steps/decode_fmllr.sh --nj 5 --cmd \"$decode_cmd\" \\\n",
    "       $tri3_step2/graph_lang_ESTER12_sat_$lg $idata_kaldi/DATA \\\n",
    "       $tri3_step2/decode_EVAL2005_sat_$lg\n",
    "done"
   ]
Abdelwahab HEBA's avatar
add new modif  
Abdelwahab HEBA committed
1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#SG build model\n",
    "local/format_lms.sh --src-dir data_last/lang_ESTER12 data_last/local_ESTER12/lm\n",
    "utils/mkgraph.sh $idata_kaldi/lang_ESTER12_new_sat_test_$lg \\\n",
    "     $tri3_step2 $tri3_step2/graph_lang_ESTER12_sat_$lg"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## DNN PART"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "exp_sat=exp_kaldi/"
   ]
Abdelwahab HEBA's avatar
add Jupyter scrips and update format_lms  
Abdelwahab HEBA committed
1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Bash",
   "language": "bash",
   "name": "bash"
  },
  "language_info": {
   "codemirror_mode": "shell",
   "file_extension": ".sh",
   "mimetype": "text/x-sh",
   "name": "bash"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}