#!/bin/bash
  
  # It's best to run the commands in this one by one.
  . ./cmd.sh
  . ./path.sh
  mfccdir=mfcc
  set -e
  rescore=true
  
  # check for kaldi_lm
  which get_word_map.pl > /dev/null
  if [ $? -ne 0 ]; then
    echo "This recipe requires installation of tools/kaldi_lm. Please run extras/kaldi_lm.sh in tools/" && exit 1;
  fi
  
  # prepare fisher data and put it under data/train_fisher
  local/fisher_data_prep.sh /export/corpora3/LDC/LDC2004T19 /export/corpora3/LDC/LDC2005T19 \
     /export/corpora3/LDC/LDC2004S13 /export/corpora3/LDC/LDC2005S13
  
  # at BUT:
  ####local/fisher_data_prep.sh /mnt/matylda6/jhu09/qpovey/FISHER/LDC2005T19 /mnt/matylda2/data/FISHER/
  
  local/swbd1_data_download.sh /export/corpora3/LDC/LDC97S62
  
  # prepare dictionary and acronym mapping list
  local/fisher_swbd_prepare_dict.sh
  
  # prepare swbd data and put it under data/train_swbd
  local/swbd1_data_prep.sh /export/corpora3/LDC/LDC97S62
  # local/swbd1_data_prep.sh /data/corpora0/LDC97S62
  # local/swbd1_data_prep.sh /mnt/matylda2/data/SWITCHBOARD_1R2
  # local/swbd1_data_prep.sh /exports/work/inf_hcrc_cstr_general/corpora/switchboard/switchboard1
  
  utils/prepare_lang.sh data/local/dict_nosp \
      "<unk>" data/local/lang_nosp data/lang_nosp
  
  # LM for swbd could be used for decoding purposes
  #fisher_opt="--fisher /scail/group/deeplearning/speech/datasets/LDC2004T19-Fisher-Transcripts"
  #local/swbd1_train_lms.sh $fisher_opt \
  #  data/local/train_swbd/text data/local/dict/lexicon.txt data/local/lm
  
  # merge two datasets into one
  mkdir -p data/train_all
  for f in spk2utt utt2spk wav.scp text segments reco2file_and_channel; do
    cat data/train_fisher/$f data/train_swbd/$f > data/train_all/$f
  done
  
  # LM for train_all
  local/fisher_train_lms.sh
  #local/fisher_create_test_lang.sh
  # Compiles G for trigram LM
  LM=data/local/lm/3gram-mincount/lm_unpruned.gz
  srilm_opts="-subset -prune-lowprobs -unk -tolower -order 3"
  utils/format_lm_sri.sh --srilm-opts "$srilm_opts" \
    data/lang_nosp $LM data/local/dict_nosp/lexicon.txt data/lang_nosp_fsh_sw1_tg
  
  LM_fg=data/local/lm/4gram-mincount/lm_unpruned.gz
  [ -f $LM_fg ] || rescore=false
  if [ $rescore ]; then
    utils/build_const_arpa_lm.sh $LM_fg data/lang_nosp data/lang_nosp_fsh_sw1_fg
  fi
  
  # Prepare Eval2000 and RT-03 test sets
  
  #local/eval2000_data_prep.sh /scail/group/deeplearning/speech/datasets/LDC2002S09/hub5e_00/ /scail/group/deeplearning/speech/datasets/LDC2002T43 || exit 1
  local/eval2000_data_prep.sh /export/corpora/LDC/LDC2002S09/hub5e_00 /export/corpora/LDC/LDC2002T43 || exit 1
  
  #local/rt03_data_prep.sh /scail/group/deeplearning/speech/datasets/rt_03 || exit 1
  local/rt03_data_prep.sh /export/corpora/LDC/LDC2007S10 || exit 1
  
  utils/fix_data_dir.sh data/train_all
  
  
  # Make MFCCs for the training set
  # spread the mfccs over various machines, as this data-set is quite large.
  if [[  $(hostname -f) ==  *.clsp.jhu.edu ]]; then
    mfcc=$(basename $mfccdir) # in case was absolute pathname (unlikely), get basename.
    utils/create_split_dir.pl /export/b{05,06,07,08}/$USER/kaldi-data/egs/fisher_swbd/s5/$mfcc/storage \
      $mfccdir/storage
  fi
  steps/make_mfcc.sh --nj 100 --cmd "$train_cmd" data/train_all exp/make_mfcc/train_all $mfccdir || exit 1;
  utils/fix_data_dir.sh data/train_all
  utils/validate_data_dir.sh data/train_all
  steps/compute_cmvn_stats.sh data/train_all exp/make_mfcc/train_all $mfccdir
  
  # subset swbd features and put them back into train_swbd in case separate training is needed
  awk -F , '{print $1}' data/train_swbd/spk2utt > data/swbd_spklist
  utils/subset_data_dir.sh --spk-list data/swbd_spklist data/train_all data/train_swbd
  steps/compute_cmvn_stats.sh data/train_swbd exp/make_mfcc/train_all $mfccdir
  
  # Make MFCCs for the test sets
  utils/fix_data_dir.sh data/eval2000
  steps/make_mfcc.sh --nj 20 --cmd "$train_cmd" data/eval2000 exp/make_mfcc/eval2000 $mfccdir || exit 1;
  steps/compute_cmvn_stats.sh data/eval2000 exp/make_mfcc/eval2000 $mfccdir
  
  utils/fix_data_dir.sh data/rt03
  steps/make_mfcc.sh --nj 20 --cmd "$train_cmd" data/rt03 exp/make_mfcc/rt03 $mfccdir || exit 1;
  steps/compute_cmvn_stats.sh data/rt03 exp/make_mfcc/rt03 $mfccdir
  
  utils/fix_data_dir.sh data/eval2000
  utils/validate_data_dir.sh data/eval2000
  
  utils/fix_data_dir.sh data/rt03
  utils/validate_data_dir.sh data/rt03
  
  n=$[`cat data/train_all/segments | wc -l`]
  utils/subset_data_dir.sh --last data/train_all $n data/train
  
  # Now-- there are 2.1 million utterances, and we want to start the monophone training
  # on relatively short utterances (easier to align), but not only the very shortest
  # ones (mostly uh-huh).  So take the 100k shortest ones, and then take 10k random
  # utterances from those. We also take these subsets from Switchboard, which has
  # more carefully hand-labeled alignments
  
  utils/subset_data_dir.sh --shortest data/train_swbd 100000 data/train_100kshort
  utils/data/remove_dup_utts.sh 10 data/train_100kshort data/train_100kshort_nodup
  utils/subset_data_dir.sh  data/train_100kshort_nodup 10000 data/train_10k_nodup
  
  utils/subset_data_dir.sh --speakers data/train_swbd 30000 data/train_30k
  utils/subset_data_dir.sh --speakers data/train_swbd 100000 data/train_100k
  
  utils/data/remove_dup_utts.sh 200 data/train_30k data/train_30k_nodup
  utils/data/remove_dup_utts.sh 200 data/train_100k data/train_100k_nodup
  utils/data/remove_dup_utts.sh 300 data/train data/train_nodup
  
  # The next commands are not necessary for the scripts to run, but increase
  # efficiency of data access by putting the mfcc's of the subset
  # in a contiguous place in a file.
  ( . ./path.sh;
    # make sure mfccdir is defined as above..
    cp data/train_10k_nodup/feats.scp{,.bak}
    copy-feats scp:data/train_10k_nodup/feats.scp  ark,scp:$mfccdir/kaldi_fish_10k_nodup.ark,$mfccdir/kaldi_fish_10k_nodup.scp \
    && cp $mfccdir/kaldi_fish_10k_nodup.scp data/train_10k_nodup/feats.scp
  )
  ( . ./path.sh;
    # make sure mfccdir is defined as above..
    cp data/train_30k_nodup/feats.scp{,.bak}
    copy-feats scp:data/train_30k_nodup/feats.scp  ark,scp:$mfccdir/kaldi_fish_30k_nodup.ark,$mfccdir/kaldi_fish_30k_nodup.scp \
    && cp $mfccdir/kaldi_fish_30k_nodup.scp data/train_30k_nodup/feats.scp
  )
  ( . ./path.sh;
    # make sure mfccdir is defined as above..
    cp data/train_100k_nodup/feats.scp{,.bak}
    copy-feats scp:data/train_100k_nodup/feats.scp  ark,scp:$mfccdir/kaldi_fish_100k_nodup.ark,$mfccdir/kaldi_fish_100k_nodup.scp \
    && cp $mfccdir/kaldi_fish_100k_nodup.scp data/train_100k_nodup/feats.scp
  )
  
  # Start training on the Switchboard subset, which has cleaner alignments
  steps/train_mono.sh --nj 3 --cmd "$train_cmd" \
    data/train_10k_nodup data/lang_nosp exp/mono0a
  
  steps/align_si.sh --nj 10 --cmd "$train_cmd" \
     data/train_30k_nodup data/lang_nosp exp/mono0a exp/mono0a_ali || exit 1;
  
  steps/train_deltas.sh --cmd "$train_cmd" \
      3200 30000 data/train_30k_nodup data/lang_nosp exp/mono0a_ali exp/tri1a || exit 1;
  #used to be 2500 20000
  (
   graph_dir=exp/tri1a/graph_nosp_fsh_sw1_tg
   utils/mkgraph.sh data/lang_nosp_fsh_sw1_tg exp/tri1a $graph_dir
   steps/decode.sh --nj 25 --cmd "$decode_cmd" --config conf/decode.config \
     $graph_dir data/eval2000 exp/tri1a/decode_eval2000_nosp_fsh_sw1_tg
   steps/decode.sh --nj 25 --cmd "$decode_cmd" --config conf/decode.config \
     $graph_dir data/rt03 exp/tri1a/decode_rt03_nosp_fsh_sw1_tg
  )&
  steps/align_si.sh --nj 10 --cmd "$train_cmd" \
     data/train_30k_nodup data/lang_nosp exp/tri1a exp/tri1a_ali || exit 1;
  
  steps/train_deltas.sh --cmd "$train_cmd" \
      3200 30000 data/train_30k_nodup data/lang_nosp exp/tri1a_ali exp/tri1b || exit 1;
  #used to be 2500 20000
  (
   graph_dir=exp/tri1b/graph_nosp_fsh_sw1_tg
   utils/mkgraph.sh data/lang_nosp_fsh_sw1_tg exp/tri1b $graph_dir
   steps/decode.sh --nj 25 --cmd "$decode_cmd" --config conf/decode.config \
     $graph_dir data/eval2000 exp/tri1b/decode_eval2000_nosp_fsh_sw1_tg
   steps/decode.sh --nj 25 --cmd "$decode_cmd" --config conf/decode.config \
     $graph_dir data/rt03 exp/tri1b/decode_rt03_nosp_fsh_sw1_tg
  )&
  steps/align_si.sh --nj 50 --cmd "$train_cmd" \
     data/train_100k_nodup data/lang_nosp exp/tri1b exp/tri1b_ali || exit 1;
  
  steps/train_deltas.sh --cmd "$train_cmd" \
      5500 90000 data/train_100k_nodup data/lang_nosp exp/tri1b_ali exp/tri2 || exit 1;
   #used to be 2500 20000 on 30k
  (
    graph_dir=exp/tri2/graph_nosp_fsh_sw1_tg
    utils/mkgraph.sh data/lang_nosp_fsh_sw1_tg exp/tri2 $graph_dir || exit 1;
    steps/decode.sh --nj 25 --cmd "$decode_cmd" --config conf/decode.config \
     $graph_dir data/eval2000 exp/tri2/decode_eval2000_nosp_fsh_sw1_tg || exit 1;
    steps/decode.sh --nj 25 --cmd "$decode_cmd" --config conf/decode.config \
     $graph_dir data/rt03 exp/tri2/decode_rt03_nosp_fsh_sw1_tg || exit 1;
  )&
  
  # Train tri3a, the last speaker-independent triphone stage,
  # on the whole Switchboard training set
  steps/align_si.sh --nj 100 --cmd "$train_cmd" \
     data/train_swbd data/lang_nosp exp/tri2 exp/tri2_ali || exit 1;
  
  steps/train_deltas.sh --cmd "$train_cmd" \
      11500 200000 data/train_swbd data/lang_nosp exp/tri2_ali exp/tri3a || exit 1;
   #used to be 2500 20000
  
  (
    graph_dir=exp/tri3a/graph_nosp_fsh_sw1_tg
    utils/mkgraph.sh data/lang_nosp_fsh_sw1_tg exp/tri3a $graph_dir || exit 1;
    steps/decode.sh --nj 25 --cmd "$decode_cmd" --config conf/decode.config \
     $graph_dir data/eval2000 exp/tri3a/decode_eval2000_nosp_fsh_sw1_tg || exit 1;
    steps/decode.sh --nj 25 --cmd "$decode_cmd" --config conf/decode.config \
     $graph_dir data/rt03 exp/tri3a/decode_rt03_nosp_fsh_sw1_tg || exit 1;
  )&
  
  # Train tri3b, which is LDA+MLLT on the whole Switchboard+Fisher training set
  steps/align_si.sh --nj 100 --cmd "$train_cmd" \
    data/train_nodup data/lang_nosp exp/tri3a exp/tri3a_ali || exit 1;
  
  steps/train_lda_mllt.sh --cmd "$train_cmd" \
     --splice-opts "--left-context=3 --right-context=3" \
     11500 400000 data/train_nodup data/lang_nosp exp/tri3a_ali exp/tri3b || exit 1;
  (
    graph_dir=exp/tri3b/graph_nosp_fsh_sw1_tg
    utils/mkgraph.sh data/lang_nosp_fsh_sw1_tg exp/tri3b $graph_dir || exit 1;
    steps/decode_fmllr.sh --nj 25 --cmd "$decode_cmd" --config conf/decode.config \
     $graph_dir data/eval2000 exp/tri3b/decode_eval2000_nosp_fsh_sw1_tg || exit 1;
    steps/decode_fmllr.sh --nj 25 --cmd "$decode_cmd" --config conf/decode.config \
     $graph_dir data/rt03 exp/tri3b/decode_rt03_nosp_fsh_sw1_tg || exit 1;
  )&
  
  steps/get_prons.sh --cmd "$train_cmd" data/train_nodup data/lang_nosp exp/tri3b
  
  utils/dict_dir_add_pronprobs.sh --max-normalize true \
    data/local/dict_nosp exp/tri3b/pron_counts_nowb.txt exp/tri3b/sil_counts_nowb.txt \
    exp/tri3b/pron_bigram_counts_nowb.txt data/local/dict
  
  utils/prepare_lang.sh data/local/dict "<unk>" data/local/lang data/lang
  
  LM=data/local/lm/3gram-mincount/lm_unpruned.gz
  srilm_opts="-subset -prune-lowprobs -unk -tolower -order 3"
  utils/format_lm_sri.sh --srilm-opts "$srilm_opts" \
    data/lang $LM data/local/dict/lexicon.txt data/lang_fsh_sw1_tg
  
  LM_fg=data/local/lm/4gram-mincount/lm_unpruned.gz
  if [ $rescore ]; then
    utils/build_const_arpa_lm.sh $LM_fg data/lang data/lang_fsh_sw1_fg
  fi
  
  (
    graph_dir=exp/tri3b/graph_fsh_sw1_tg
    utils/mkgraph.sh data/lang_fsh_sw1_tg exp/tri3b $graph_dir || exit 1;
    steps/decode_fmllr.sh --nj 25 --cmd "$decode_cmd" --config conf/decode.config \
     $graph_dir data/eval2000 exp/tri3b/decode_eval2000_fsh_sw1_tg || exit 1;
    steps/decode_fmllr.sh --nj 25 --cmd "$decode_cmd" --config conf/decode.config \
     $graph_dir data/rt03 exp/tri3b/decode_rt03_fsh_sw1_tg || exit 1;
  )&
  
  # Next we'll use fMLLR and train with SAT (i.e. on
  # fMLLR features)
  
  steps/align_fmllr.sh --nj 100 --cmd "$train_cmd" \
    data/train_nodup data/lang exp/tri3b exp/tri3b_ali || exit 1;
  
  steps/train_sat.sh  --cmd "$train_cmd" \
    11500 800000 data/train_nodup data/lang exp/tri3b_ali  exp/tri4a || exit 1;
  
  (
    graph_dir=exp/tri4a/graph_fsh_sw1_tg
    utils/mkgraph.sh data/lang_fsh_sw1_tg exp/tri4a $graph_dir
    steps/decode_fmllr.sh --nj 25 --cmd "$decode_cmd" --config conf/decode.config \
     $graph_dir data/eval2000 exp/tri4a/decode_eval2000_fsh_sw1_tg || exit 1;
    steps/decode_fmllr.sh --nj 25 --cmd "$decode_cmd" --config conf/decode.config \
     $graph_dir data/rt03 exp/tri4a/decode_rt03_fsh_sw1_tg || exit 1;
  )&
  wait
  
  if [ $rescore ]; then
    steps/lmrescore_const_arpa.sh --cmd "$decode_cmd" \
      data/lang_fsh_sw1_{tg,fg} data/eval2000 \
      exp/tri4a/decode_eval2000_fsh_sw1_{tg,fg}
    steps/lmrescore_const_arpa.sh --cmd "$decode_cmd" \
      data/lang_fsh_sw1_{tg,fg} data/rt03 \
      exp/tri4a/decode_rt03_fsh_sw1_{tg,fg}
  fi
  
  steps/align_fmllr.sh --nj 100 --cmd "$train_cmd" \
    data/train_nodup data/lang exp/tri4a exp/tri4a_ali || exit 1;
  
  steps/train_sat.sh  --cmd "$train_cmd" \
    11500 1600000 data/train_nodup data/lang exp/tri4a_ali  exp/tri5a || exit 1;
  
  (
    graph_dir=exp/tri5a/graph_fsh_sw1_tg
    utils/mkgraph.sh data/lang_fsh_sw1_tg exp/tri5a $graph_dir
    steps/decode_fmllr.sh --nj 25 --cmd "$decode_cmd" --config conf/decode.config \
     $graph_dir data/eval2000 exp/tri5a/decode_eval2000_fsh_sw1_tg || exit 1;
    steps/decode_fmllr.sh --nj 25 --cmd "$decode_cmd" --config conf/decode.config \
     $graph_dir data/rt03 exp/tri5a/decode_rt03_fsh_sw1_tg || exit 1;
  )&
  wait
  
  if [ $rescore ]; then
    steps/lmrescore_const_arpa.sh --cmd "$decode_cmd" \
      data/lang_fsh_sw1_{tg,fg} data/eval2000 \
      exp/tri5a/decode_eval2000_fsh_sw1_{tg,fg}
    steps/lmrescore_const_arpa.sh --cmd "$decode_cmd" \
      data/lang_fsh_sw1_{tg,fg} data/rt03 \
      exp/tri5a/decode_rt03_fsh_sw1_{tg,fg}
  fi
  
  hours=$(awk '{x += $4 - $3;} END{print x/3600;}' <data/train_fisher/segments)
  ! [ $hours == 1915 ] && echo "$0: expected 1915 hours of data, got $hours hours, please check." && exit 1;
  
  # at BUT:
  ####local/fisher_data_prep.sh /mnt/matylda6/jhu09/qpovey/FISHER/LDC2005T19 /mnt/matylda2/data/FISHER/
  
  steps/align_fmllr.sh --nj 100 --cmd "$train_cmd" \
    data/train_nodup data/lang exp/tri5a exp/tri5a_ali || exit 1;
  
  
  steps/train_sat.sh  --cmd "$train_cmd" \
    11500 3200000 data/train_nodup data/lang exp/tri5a_ali  exp/tri6a || exit 1;
  
  (
    graph_dir=exp/tri6a/graph_fsh_sw1_tg
    utils/mkgraph.sh data/lang_fsh_sw1_tg exp/tri6a $graph_dir
    steps/decode_fmllr.sh --nj 25 --cmd "$decode_cmd" --config conf/decode.config \
     $graph_dir data/eval2000 exp/tri6a/decode_eval2000_fsh_sw1_tg || exit 1;
    steps/decode_fmllr.sh --nj 25 --cmd "$decode_cmd" --config conf/decode.config \
     $graph_dir data/rt03 exp/tri6a/decode_rt03_fsh_sw1_tg || exit 1;
  )&
  wait
  if [ $rescore ]; then
    steps/lmrescore_const_arpa.sh --cmd "$decode_cmd" \
      data/lang_fsh_sw1_{tg,fg} data/eval2000 \
      exp/tri6a/decode_eval2000_fsh_sw1_{tg,fg}
    steps/lmrescore_const_arpa.sh --cmd "$decode_cmd" \
      data/lang_fsh_sw1_{tg,fg} data/rt03 \
      exp/tri6a/decode_rt03_fsh_sw1_{tg,fg}
  fi
  
  # Optional tri6a alignment for further training purposes
  
  #steps/align_fmllr.sh --nj 200 --cmd "$train_cmd" \
  #  data/train_nodup data/lang exp/tri6a exp/tri6a_ali || exit 1;
  
  # The following is the current online-nnet2 recipe, with "multi-splice".
  # local/online/run_nnet2_ms.sh
  local/online/run_nnet2_ms.sh