[examples] update units in examples (wenet-e2e#1217)

Mddct · Jun 11, 2022 · 7083e6f · 7083e6f
1 parent 46b284a
commit 7083e6f
Show file tree

Hide file tree

Showing 9 changed files with 19 additions and 11 deletions.
diff --git a/.github/workflows/docker_image.yml b/.github/workflows/docker_image.yml
@@ -49,7 +49,7 @@ jobs:
           --port 10086 \
           --chunk_size 16 \
           --model_path \$model/final.zip \
-          --dict_path \$model/words.txt 2>&1 | tee server.log
+          --unit_path \$model/units.txt 2>&1 | tee server.log
         EOF
 
         id=$(docker create ubuntu:latest)

diff --git a/docs/context.md b/docs/context.md
@@ -132,7 +132,7 @@ model_dir=docker_resource/model
     --model_path $model_dir/final.zip \
     --context_path $context_path \
     --context_score 3 \
-    --dict_path $model_dir/words.txt 2>&1 | tee log.txt
+    --unit_path $model_dir/units.txt 2>&1 | tee log.txt
 ```
 
 
diff --git a/docs/lm.md b/docs/lm.md
@@ -100,6 +100,7 @@ tools/fst/make_tlg.sh data/local/lm data/local/lang data/lang_test || exit 1;
     --beam 15.0 --lattice_beam 7.5 --max_active 7000 \
     --blank_skip_thresh 0.98 --ctc_weight 0.5 --rescoring_weight 1.0 \
     --fst_path data/lang_test/TLG.fst \
+    --dict_path data/lang_test/words.txt \
     data/test/wav.scp data/test/text $dir/final.zip \
-    data/lang_test/words.txt $dir/lm_with_runtime
+    data/lang_test/units.txt $dir/lm_with_runtime
 ```
diff --git a/docs/tutorial_librispeech.md b/docs/tutorial_librispeech.md
@@ -422,7 +422,8 @@ if [ ${stage} -le 7 ] && [ ${stop_stage} -ge 7 ]; then
       --beam 10.0 --lattice_beam 5 --max_active 7000 --blank_skip_thresh 0.98 \
       --ctc_weight 0.5 --rescoring_weight 1.0 --acoustic_scale 1.2 \
       --fst_path $fst_dir/TLG.fst \
-      data/$test/wav.scp data/$test/text $dir/final.zip $fst_dir/words.txt \
+      --dict_path $fst_dir/words.txt \
+      data/$test/wav.scp data/$test/text $dir/final.zip $fst_dir/units.txt \
       $dir/lm_with_runtime_${test}
     tail $dir/lm_with_runtime_${test}/wer
   done

diff --git a/examples/aishell/s0/run.sh b/examples/aishell/s0/run.sh
@@ -233,8 +233,9 @@ if [ ${stage} -le 7 ] && [ ${stop_stage} -ge 7 ]; then
     --blank_skip_thresh 0.98 --ctc_weight 0.5 --rescoring_weight 1.0 \
     --chunk_size $chunk_size \
     --fst_path data/lang_test/TLG.fst \
+    --dict_path data/lang_test/words.fst \
     data/test/wav.scp data/test/text $dir/final.zip \
-    data/lang_test/words.txt $dir/lm_with_runtime
+    data/lang_test/units.txt $dir/lm_with_runtime
   # Please see $dir/lm_with_runtime for wer
 fi
 

diff --git a/examples/aishell/s0_deprecated/run.sh b/examples/aishell/s0_deprecated/run.sh
@@ -239,8 +239,9 @@ if [ ${stage} -le 7 ] && [ ${stop_stage} -ge 7 ]; then
         --blank_skip_thresh 0.98 --ctc_weight 0.5 --rescoring_weight 1.0 \
         --reverse_weight $reverse_weight --chunk_size $chunk_size \
         --fst_path data/lang_test/TLG.fst \
+        --dict_path data/lang_test/words.txt \
         data/test/wav.scp data/test/text $dir/final.zip \
-        data/lang_test/words.txt $dir/lm_with_runtime
+        data/lang_test/units.txt $dir/lm_with_runtime
     # See $dir/lm_with_runtime for wer
 fi
 

diff --git a/examples/aishell2/s0/run.sh b/examples/aishell2/s0/run.sh
@@ -224,7 +224,8 @@ if [ ${stage} -le 7 ] && [ ${stop_stage} -ge 7 ]; then
       --beam 15.0 --lattice_beam 7.5 --max_active 7000 --blank_skip_thresh 0.98 \
       --ctc_weight 0.3 --rescoring_weight 1.0 --reverse_weight $reverse_weight\
       --fst_path data/lang_test/TLG.fst \
-      data/test/wav.scp data/test/text $dir/final.zip data/lang_test/words.txt \
+      --dict_path data/lang_test/words.txt \
+      data/test/wav.scp data/test/text $dir/final.zip data/lang_test/units.txt \
       $dir/lm_with_runtime
   # See $dir/lm_with_runtime for wer
   tail $dir/lm_with_runtime/wer

diff --git a/examples/librispeech/s0/run.sh b/examples/librispeech/s0/run.sh
@@ -273,7 +273,8 @@ if [ ${stage} -le 7 ] && [ ${stop_stage} -ge 7 ]; then
       --beam 10.0 --lattice_beam 5 --max_active 7000 --blank_skip_thresh 0.98 \
       --ctc_weight 0.5 --rescoring_weight 1.0 --acoustic_scale 1.2 \
       --fst_path $fst_dir/TLG.fst \
-      data/$test/wav.scp data/$test/text $dir/final.zip $fst_dir/words.txt \
+      --dict_path $fst_dir/words.txt \
+      data/$test/wav.scp data/$test/text $dir/final.zip $fst_dir/units.txt \
       $dir/lm_with_runtime_${test}
     tail $dir/lm_with_runtime_${test}/wer
   done

diff --git a/tools/decode.sh b/tools/decode.sh
@@ -13,6 +13,7 @@ reverse_weight=0.0
 rescoring_weight=1.0
 # For CTC WFST based decoding
 fst_path=
+dict_path=
 acoustic_scale=1.0
 beam=15.0
 lattice_beam=12.0
@@ -23,7 +24,7 @@ length_penalty=0.0
 
 . tools/parse_options.sh || exit 1;
 if [ $# != 5 ]; then
-  echo "Usage: $0 [options] <wav.scp> <label_file> <model_file> <dict_file> <output_dir>"
+  echo "Usage: $0 [options] <wav.scp> <label_file> <model_file> <unit_file> <output_dir>"
   exit 1;
 fi
 
@@ -35,7 +36,7 @@ fi
 scp=$1
 label_file=$2
 model_file=$3
-dict_file=$4
+unit_file=$4
 dir=$5
 
 mkdir -p $dir/split${nj}
@@ -52,6 +53,7 @@ wfst_decode_opts=
 if [ ! -z $fst_path ]; then
   wfst_decode_opts="--fst_path $fst_path"
   wfst_decode_opts="$wfst_decode_opts --beam $beam"
+  wfst_decode_opts="$wfst_decode_opts --dict_path $dict_path"
   wfst_decode_opts="$wfst_decode_opts --lattice_beam $lattice_beam"
   wfst_decode_opts="$wfst_decode_opts --max_active $max_active"
   wfst_decode_opts="$wfst_decode_opts --min_active $min_active"
@@ -69,7 +71,7 @@ for n in $(seq ${nj}); do
      --chunk_size $chunk_size \
      --wav_scp ${dir}/split${nj}/wav.${n}.scp \
      --model_path $model_file \
-     --dict_path $dict_file \
+     --unit_path $unit_file \
      $wfst_decode_opts \
      --result ${dir}/split${nj}/${n}.text &> ${dir}/split${nj}/${n}.log
 } &