icefall/egs/mucs/ASR/prepare.sh
2023-05-01 17:32:00 +05:30

233 lines
6.0 KiB
Bash
Executable File

#!/usr/bin/env bash
# fix segmentation fault reported in https://github.com/k2-fsa/icefall/issues/674
export PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION=python
set -eou pipefail
nj=60
stage=6
stop_stage=9
# We assume dl_dir (download dir) contains the following
# directories and files. If not, they will be downloaded
# by this script automatically.
#
# - $dl_dir/hi-en
dl_dir=$PWD/download
espnet_path=/home/wtc7/espnet/egs2/MUCS/asr1/data/hi-en/
. shared/parse_options.sh || exit 1
# vocab size for sentence piece models.
# It will generate data/lang_bpe_xxx,
# data/lang_bpe_yyy
vocab_size=400
# All files generated by this script are saved in "data".
# You can safely remove "data" and rerun this script to regenerate it.
mkdir -p data
log() {
# This function is from espnet
local fname=${BASH_SOURCE[1]##*/}
echo -e "$(date '+%Y-%m-%d %H:%M:%S') (${fname}:${BASH_LINENO[0]}:${FUNCNAME[1]}) $*"
}
log "dl_dir: $dl_dir"
if [ $stage -le -1 ] && [ $stop_stage -ge -1 ]; then
log "Stage -1: prepare LM files"
mkdir -p $dl_dir/lm
if [ ! -e $dl_dir/lm/.done ]; then
./local/prepare_lm_files.py --out-dir=$dl_dir/lm --data-path=$espnet_path --mode="train"
touch $dl_dir/lm/.done
fi
fi
if [ $stage -le 0 ] && [ $stop_stage -ge 0 ]; then
log "Stage 0: Download data"
fi
if [ $stage -le 1 ] && [ $stop_stage -ge 1 ]; then
log "Stage 1: Prepare MUCS manifest"
# We assume that you have downloaded the MUCS corpus
# to $dl_dir/
mkdir -p data/manifests
if [ ! -e data/manifests/.mucs.done ]; then
# generate lhotse manifests from kaldi style files
./local/prepare_manifest.py "$espnet_path" $nj data/manifests
touch data/manifests/.mucs.done
fi
fi
if [ $stage -le 3 ] && [ $stop_stage -ge 3 ]; then
log "Stage 3: Compute fbank for mucs"
mkdir -p data/fbank
if [ ! -e data/fbank/.mucs.done ]; then
./local/compute_fbank_mucs.py
touch data/fbank/.mucs.done
fi
# exit
if [ ! -e data/fbank/.mucs-validated.done ]; then
log "Validating data/fbank for mucs"
parts=(
train
test
dev
)
for part in ${parts[@]}; do
python3 ./local/validate_manifest.py \
data/fbank/mucs_cuts_${part}.jsonl.gz
done
touch data/fbank/.mucs-validated.done
fi
fi
if [ $stage -le 5 ] && [ $stop_stage -ge 5 ]; then
log "Stage 5: Prepare phone based lang"
lang_dir=data/lang_phone
mkdir -p $lang_dir
(echo '!SIL SIL'; echo '<SPOKEN_NOISE> SPN'; echo '<UNK> SPN'; ) |
cat - $dl_dir/lm/mucs_lexicon.txt |
sort | uniq > $lang_dir/lexicon.txt
if [ ! -f $lang_dir/L_disambig.pt ]; then
./local/prepare_lang.py --lang-dir $lang_dir
fi
if [ ! -f $lang_dir/L.fst ]; then
log "Converting L.pt to L.fst"
./shared/convert-k2-to-openfst.py \
--olabels aux_labels \
$lang_dir/L.pt \
$lang_dir/L.fst
fi
if [ ! -f $lang_dir/L_disambig.fst ]; then
log "Converting L_disambig.pt to L_disambig.fst"
./shared/convert-k2-to-openfst.py \
--olabels aux_labels \
$lang_dir/L_disambig.pt \
$lang_dir/disambig_L.fst
fi
fi
if [ $stage -le 6 ] && [ $stop_stage -ge 6 ]; then
log "Stage 6: Prepare BPE based lang"
lang_dir=data/lang_bpe_${vocab_size}
mkdir -p $lang_dir
# We reuse words.txt from phone based lexicon
# so that the two can share G.pt later.
cp data/lang_phone/words.txt $lang_dir
if [ ! -f $lang_dir/transcript_words.txt ]; then
log "Generate data for BPE training"
cp download/lm/mucs_vocab_text.txt $lang_dir/transcript_words.txt
fi
if [ ! -f $lang_dir/bpe.model ]; then
./local/train_bpe_model.py \
--lang-dir $lang_dir \
--vocab-size $vocab_size \
--transcript $lang_dir/transcript_words.txt
fi
if [ ! -f $lang_dir/L_disambig.pt ]; then
./local/prepare_lang_bpe.py --lang-dir $lang_dir
log "Validating $lang_dir/lexicon.txt"
./local/validate_bpe_lexicon.py \
--lexicon $lang_dir/lexicon.txt \
--bpe-model $lang_dir/bpe.model
fi
if [ ! -f $lang_dir/L.fst ]; then
log "Converting L.pt to L.fst"
./shared/convert-k2-to-openfst.py \
--olabels aux_labels \
$lang_dir/L.pt \
$lang_dir/L.fst
fi
if [ ! -f $lang_dir/L_disambig.fst ]; then
log "Converting L_disambig.pt to L_disambig.fst"
./shared/convert-k2-to-openfst.py \
--olabels aux_labels \
$lang_dir/L_disambig.pt \
$lang_dir/L_disambig.fst
fi
fi
if [ $stage -le 7 ] && [ $stop_stage -ge 7 ]; then
log "Stage 7: Train LM from training data"
lang_dir=data/lang_bpe_${vocab_size}
if [ ! -f $lang_dir/lm_3.arpa ]; then
./shared/make_kn_lm.py \
-ngram-order 3 \
-text $lang_dir/transcript_words.txt \
-lm $lang_dir/lm_3.arpa
fi
if [ ! -f $lang_dir/lm_4.arpa ]; then
./shared/make_kn_lm.py \
-ngram-order 4 \
-text $lang_dir/transcript_words.txt \
-lm $lang_dir/lm_4.arpa
fi
fi
if [ $stage -le 8 ] && [ $stop_stage -ge 8 ]; then
log "Stage 8: Prepare G"
# We assume you have install kaldilm, if not, please install
# it using: pip install kaldilm
mkdir -p data/lm
if [ ! -f data/lm/G_3_gram.fst.txt ]; then
# It is used in building HLG
python3 -m kaldilm \
--read-symbol-table="data/lang_phone/words.txt" \
--disambig-symbol='#0' \
--max-order=3 \
data/lang_bpe_${vocab_size}/lm_3.arpa > data/lm/G_3_gram.fst.txt
fi
if [ ! -f data/lm/G_4_gram.fst.txt ]; then
# It is used in building HLG
python3 -m kaldilm \
--read-symbol-table="data/lang_phone/words.txt" \
--disambig-symbol='#0' \
--max-order=3 \
data/lang_bpe_${vocab_size}/lm_4.arpa > data/lm/G_4_gram.fst.txt
fi
fi
if [ $stage -le 9 ] && [ $stop_stage -ge 9 ]; then
log "Stage 9: Compile HLG"
# ./local/compile_hlg.py --lang-dir data/lang_phone
# Note If ./local/compile_hlg.py throws OOM,
# please switch to the following command
#
# ./local/compile_hlg_using_openfst.py --lang-dir data/lang_phone
lang_dir=data/lang_bpe_${vocab_size}
./local/compile_hlg.py --lang-dir $lang_dir
fi