Separate number of jobs for decoding and fix another dictionary issue

git-svn-id: https://svn.code.sf.net/p/kaldi/code/sandbox/nshmyrev@3851 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8
batermj · Apr 6, 2014 · 803d345 · 803d345
1 parent 02cc48b
commit 803d345
Show file tree

Hide file tree

Showing 2 changed files with 8 additions and 5 deletions.
diff --git a/egs/tedlium/s5/local/dict_prep.sh b/egs/tedlium/s5/local/dict_prep.sh
@@ -4,9 +4,11 @@ mkdir -p $dir
 
 srcdict=db/TEDLIUM_release1/TEDLIUM.150K.dic
 
+# Join dicts and fix some troubles
 cat $srcdict db/extra.dic | LANG= LC_ALL= sort | sed 's:([0-9])::g' |
    grep -vw "ei" |
-   grep -vw "erj" > $dir/lexicon_words.txt 
+   grep -vw "erj" |
+   grep -v "text2pho.sh" > $dir/lexicon_words.txt 
 
 cat $dir/lexicon_words.txt | awk '{ for(n=2;n<=NF;n++){ phones[$n] = 1; }} END{for (p in phones) print p;}' | \
   grep -v SIL | sort > $dir/nonsilence_phones.txt  

diff --git a/egs/tedlium/s5/run.sh b/egs/tedlium/s5/run.sh
@@ -4,6 +4,7 @@
 . path.sh
 
 nj=8
+decode_nj=4
 
 # Data prep
 
@@ -37,7 +38,7 @@ steps/train_deltas.sh --cmd "$train_cmd" \
 
 utils/mkgraph.sh data/lang_test exp/tri1 exp/tri1/graph
 
-steps/decode.sh --nj $nj --cmd "$decode_cmd" \
+steps/decode.sh --nj $decode_nj --cmd "$decode_cmd" \
   exp/tri1/graph data/test exp/tri1/decode
 
 steps/align_si.sh --nj $nj --cmd "$train_cmd" \
@@ -48,7 +49,7 @@ steps/train_lda_mllt.sh --cmd "$train_cmd" \
 
 utils/mkgraph.sh data/lang_test exp/tri2 exp/tri2/graph
 
-steps/decode.sh --nj $nj --cmd "$decode_cmd" \
+steps/decode.sh --nj $decode_nj --cmd "$decode_cmd" \
    exp/tri2/graph data/test exp/tri2/decode
 
 steps/align_si.sh --nj $nj --cmd "$train_cmd" \
@@ -59,7 +60,7 @@ steps/train_sat.sh --cmd "$train_cmd" \
 
 utils/mkgraph.sh data/lang_test exp/tri3 exp/tri3/graph
 
-steps/decode_fmllr.sh --nj $nj --cmd "$decode_cmd" \
+steps/decode_fmllr.sh --nj $decode_nj --cmd "$decode_cmd" \
    exp/tri3/graph data/test exp/tri3/decode
 
 steps/align_fmllr.sh --nj $nj --cmd "$train_cmd" \
@@ -73,6 +74,6 @@ steps/train_mmi.sh --cmd "$train_cmd" --boost 0.1 \
   exp/tri3_mmi_b0.1
 
 for iter in 3 4; do
-steps/decode.sh --nj $nj --cmd "$decode_cmd" --iter $iter \
+steps/decode.sh --nj $decode_nj --cmd "$decode_cmd" --iter $iter \
    exp/tri3/graph data/test exp/tri3_mmi_b0.1/decode_it$iter
 done