#!/bin/bash -u
  
  # Copyright 2012  Arnab Ghoshal
  
  # Licensed under the Apache License, Version 2.0 (the "License");
  # you may not use this file except in compliance with the License.
  # You may obtain a copy of the License at
  #
  #  http://www.apache.org/licenses/LICENSE-2.0
  #
  # THIS CODE IS PROVIDED *AS IS* BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
  # KIND, EITHER EXPRESS OR IMPLIED, INCLUDING WITHOUT LIMITATION ANY IMPLIED
  # WARRANTIES OR CONDITIONS OF TITLE, FITNESS FOR A PARTICULAR PURPOSE,
  # MERCHANTABLITY OR NON-INFRINGEMENT.
  # See the Apache 2 License for the specific language governing permissions and
  # limitations under the License.
  
  exit 1;
  # This script shows the steps needed to build a recognizer for certain languages
  # (currently German, Portuguese, Spanish, and Swedish) of the GlobalPhone 
  # corpus. But executing this script by itself will fail. For example, you need
  # to provide the location of the corpus. Also, you may want to run the 
  # recognition network creation steps on a machine with enough memory.
  
  # !!! NOTE: The current recipe assumes that you have pre-build LMs. 
  
  # The following data preparation step actually converts the audio files from 
  # shorten to WAV to take out the empty files and those with compression errors. 
  # So set WORKDIR to someplace with enough disk space. That is where MFCCs will 
  # get created, as well as the FST versions of LMs.
  WORKDIR=/mnt/matylda6/jhu09/qpovey/temp_gp
  cp -r conf local utils steps install.sh path.sh $WORKDIR
  cd $WORKDIR
  # INSTALLING REQUIRED TOOLS:
  #  This recipe requires shorten and sox (we use shorten 3.6.1 and sox 14.3.2).
  #  If you don't have them, use the install.sh script to install them.
  ( which shorten >&/dev/null && which sox >&/dev/null && \
    ehco "shorten and sox found: you may want to edit the path.sh file." ) || \
    { echo "shorten and/or sox not found on PATH. Installing..."; 
      install.sh }
  
  local/gp_data_prep.sh --config-dir=$PWD/conf --corpus-dir=/mnt/matylda2/data/GLOBALPHONE --lm-dir=/path/to/lms --work-dir=$WORKDIR
  # On Eddie: local/gp_data_prep.sh --config-dir=$PWD/conf --corpus-dir=$PWD/corpus --lm-dir=$PWD/corpus/language_models --work-dir=$PWD
  
  local/gp_format_data.sh --hmm-proto=conf/topo.proto --work-dir=$PWD
  
  # Now make MFCC features.
  cmd="qsub -q all.q@@blade"  # Configured for BUT: remove the option if no queue.
  for LCODE in GE PO SP SW; do
    mfccdir=$WORKDIR/data/$LCODE/MFCC
    for x in train dev eval; do
      steps/make_mfcc.sh --num-jobs 6 --qcmd "$cmd" data/$LCODE/$x \
        exp/$LCODE/make_mfcc/$x $mfccdir
    done
  done
  
  # you can change these commands to just run.pl to make them run
  # locally, but in that case you should change the num-jobs to
  # the #cpus on your machine or fewer.
  decode_cmd="queue.pl -q all.q@@blade -l ram_free=1200M,mem_free=1200M"
  train_cmd="queue.pl -q all.q@@blade -l ram_free=700M,mem_free=700M"
  
  for LCODE in GE PO SP SW; do
    steps/train_mono.sh --num-jobs 10 --qcmd "$train_cmd" \
      data/$LCODE/train data/$LCODE/lang exp/$LCODE/mono
    # The following 3 commands will not run as written, since the LM directories
    # will be different across sites. Edit the 'lang_test' to match what is 
    # available
    utils/mkgraph.sh data/$LCODE/lang_test exp/$LCODE/mono \
      exp/$LCODE/mono/graph
    utils/decode.sh --qcmd "$decode_cmd" steps/decode_deltas.sh \
      exp/$LCODE/mono/graph data/$LCODE/dev exp/$LCODE/mono/decode_dev
    utils/decode.sh --qcmd "$decode_cmd" steps/decode_deltas.sh \
      exp/$LCODE/mono/graph data/$LCODE/eval exp/$LCODE/mono/decode_eval
  done
  
  
  # This queue option will be supplied to all alignment
  # and training scripts.  Note: you have to supply the same num-jobs
  # to the alignment and training scripts, as the archives are split
  # up in this way.
  
  for LCODE in GE PO SP SW; do
    steps/align_deltas.sh --num-jobs 10 --qcmd "$train_cmd" \
      data/$LCODE/train data/$LCODE/lang exp/$LCODE/mono exp/$LCODE/mono_ali
  
    steps/train_deltas.sh --num-jobs 10 --qcmd "$train_cmd" \
      2000 10000 data/$LCODE/train data/$LCODE/lang exp/$LCODE/mono_ali \
      exp/$LCODE/tri1
  
    # Like with the monophone systems, the following 3 commands will not run.
    # Edit the 'lang_test' to match what is available.
    utils/mkgraph.sh data/$LCODE/lang_test exp/$LCODE/tri1 exp/$LCODE/tri1/graph
    utils/decode.sh --qcmd "$decode_cmd" steps/decode_deltas.sh \
      exp/$LCODE/tri1/graph data/$LCODE/dev exp/$LCODE/tri1/decode_dev
    utils/decode.sh --qcmd "$decode_cmd" steps/decode_deltas.sh \
      exp/$LCODE/tri1/graph data/$LCODE/eval exp/$LCODE/tri1/decode
  
  done