Commit f72d2a7d authored by Pawel Swietojanski's avatar Pawel Swietojanski
Browse files

sandbox/pawel: moved wordlist to openslr and altered scripts accordingly

git-svn-id: https://svn.code.sf.net/p/kaldi/code/sandbox/pawel@4271 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8
parent 347f241e
...@@ -25,8 +25,9 @@ grep -v ';;;' $dir/cmudict/cmudict.0.7a | \ ...@@ -25,8 +25,9 @@ grep -v ';;;' $dir/cmudict/cmudict.0.7a | \
#cat eddie_data/rt09.ami.ihmtrain09.v3.dct | sort > $dir/lexicon1_raw_nosil.txt #cat eddie_data/rt09.ami.ihmtrain09.v3.dct | sort > $dir/lexicon1_raw_nosil.txt
# limit the vocabulary to the predefined 50k words # limit the vocabulary to the predefined 50k words
gunzip local/wordlist.50k.gz wget -nv -O $dir/wordlist.50k.gz http://www.openslr.org/resources/9/wordlist.50k.gz
join $dir/lexicon1_raw_nosil.txt local/wordlist.50k > $dir/lexicon1_raw_nosil_50k.txt gunzip -c $dir/wordlist.50k.gz > $dir/wordlist.50k
join $dir/lexicon1_raw_nosil.txt $dir/wordlist.50k > $dir/lexicon1_raw_nosil_50k.txt
# Add prons for laughter, noise, oov # Add prons for laughter, noise, oov
for w in `grep -v sil $dir/silence_phones.txt`; do for w in `grep -v sil $dir/silence_phones.txt`; do
......
...@@ -37,7 +37,7 @@ else ...@@ -37,7 +37,7 @@ else
echo "$0. Java not found. Will download exported version of transcripts." echo "$0. Java not found. Will download exported version of transcripts."
annots=ami_manual_annotations_v1.6.1_export annots=ami_manual_annotations_v1.6.1_export
wget -O $wdir/$annots.gzip http://groups.inf.ed.ac.uk/ami/AMICorpusAnnotations/$annots.gzip wget -O $wdir/$annots.gzip http://groups.inf.ed.ac.uk/ami/AMICorpusAnnotations/$annots.gzip
gunzip -c $adir/${annots}.gzip > $wdir/transcripts0 gunzip -c $wdir/${annots}.gzip > $wdir/transcripts0
fi fi
#remove NXT logs dumped to stdio #remove NXT logs dumped to stdio
......
...@@ -5,7 +5,6 @@ KALDI_ROOT=/gpfs/scratch/s1136550/kaldi-code ...@@ -5,7 +5,6 @@ KALDI_ROOT=/gpfs/scratch/s1136550/kaldi-code
#KALDI_ROOT=/disk/data1/software/kaldi-trunk-atlas #KALDI_ROOT=/disk/data1/software/kaldi-trunk-atlas
#KALDI_ROOT=/disk/data1/pbell1/software/kaldi-trunk-mkl/ #KALDI_ROOT=/disk/data1/pbell1/software/kaldi-trunk-mkl/
KALDISRC=$KALDI_ROOT/src KALDISRC=$KALDI_ROOT/src
KALDIBIN=$KALDISRC/bin:$KALDISRC/featbin:$KALDISRC/fgmmbin:$KALDISRC/fstbin KALDIBIN=$KALDISRC/bin:$KALDISRC/featbin:$KALDISRC/fgmmbin:$KALDISRC/fstbin
KALDIBIN=$KALDIBIN:$KALDISRC/gmmbin:$KALDISRC/latbin:$KALDISRC/nnetbin KALDIBIN=$KALDIBIN:$KALDISRC/gmmbin:$KALDISRC/latbin:$KALDISRC/nnetbin
...@@ -14,8 +13,8 @@ KALDIBIN=$KALDIBIN:$KALDISRC/sgmmbin:$KALDISRC/tiedbin ...@@ -14,8 +13,8 @@ KALDIBIN=$KALDIBIN:$KALDISRC/sgmmbin:$KALDISRC/tiedbin
FSTBIN=$KALDI_ROOT/tools/openfst/bin FSTBIN=$KALDI_ROOT/tools/openfst/bin
LMBIN=$KALDI_ROOT/tools/irstlm/bin LMBIN=$KALDI_ROOT/tools/irstlm/bin
SRILM=$KALDI_ROOT/tools/srilm/bin/i686-m64 SRILM=$KALDI_ROOT/tools/srilm/bin/i686-m64
#BEAMFORMIT=$KALDI_ROOT/tools/BeamformIt-3.51 BEAMFORMIT=$KALDI_ROOT/tools/BeamformIt-3.51
BEAMFORMIT=/disk/data1/s1136550/BeamformIt-3.51 #BEAMFORMIT=/disk/data1/s1136550/BeamformIt-3.51
[ -d $PWD/local ] || { echo "Error: 'local' subdirectory not found."; } [ -d $PWD/local ] || { echo "Error: 'local' subdirectory not found."; }
[ -d $PWD/utils ] || { echo "Error: 'utils' subdirectory not found."; } [ -d $PWD/utils ] || { echo "Error: 'utils' subdirectory not found."; }
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment