#!/bin/bash



#Author: Ruchao Fan

#2017.11.1     Training acoustic model and decode with phoneme-level bigram

#2018.4.30     Replace the h5py with ark and simplify the data_loader.py

#2019.12.20    Update to pytorch1.2 and python3.7



KALDI_ROOT=./kaldi



. $KALDI_ROOT/tools/config/common_path.sh

export LC_ALL=C



stage=0



timit_dir='../data'

phoneme_map='60-39'

feat_dir='data'                            #dir to save feature

feat_type='fbank'                          #fbank, mfcc, spectrogram

config_file='conf/ctc_config.yaml'



if [ ! -z $1 ]; then

    stage=$1

fi



if [ $stage -le 0 ]; then

    echo "Step 0: Data Preparation ..."

    local/timit_data_prep.sh $timit_dir $phoneme_map || exit 1;

    python3 steps/get_model_units.py $feat_dir/train/phn_text

fi



if [ $stage -le 1 ]; then

    echo "Step 1: Feature Extraction..."

    steps/make_feat.sh $feat_type $feat_dir || exit 1;

fi