update repo

egs/aishell/conformer/run.sh

@@ -88,11 +88,11 @@ if [ ${stage} -le 1 ] && [ ${stop_stage} -ge 1 ]; then
     utils/compute_cmvn.sh --fbankdir ${feats_dir}/data/${train_set} --cmd "$train_cmd" --nj $nj --feats_dim ${feats_dim} --config_file "$asr_config" --scale 1.0
 fi
 
-token_list=${feats_dir}/data/${lang}_token_list/char/tokens.txt
+token_list=${feats_dir}/data/${lang}_token_list/$token_type/tokens.txt
 echo "dictionary: ${token_list}"
 if [ ${stage} -le 2 ] && [ ${stop_stage} -ge 2 ]; then
     echo "stage 2: Dictionary Preparation"
-    mkdir -p ${feats_dir}/data/${lang}_token_list/char/
+    mkdir -p ${feats_dir}/data/${lang}_token_list/$token_type/
 
     echo "make a dictionary"
     echo "<blank>" > ${token_list}
@@ -130,7 +130,7 @@ if [ ${stage} -le 4 ] && [ ${stop_stage} -ge 4 ]; then
                 --task_name asr \
                 --gpu_id $gpu_id \
                 --use_preprocessor true \
-                --token_type char \
+                --token_type $token_type \
                 --token_list $token_list \
                 --data_dir ${feats_dir}/data \
                 --train_set ${train_set} \

egs/aishell/data2vec_paraformer_finetune/run.sh

@@ -91,11 +91,11 @@ if [ ${stage} -le 1 ] && [ ${stop_stage} -ge 1 ]; then
     utils/compute_cmvn.sh --fbankdir ${feats_dir}/data/${train_set} --cmd "$train_cmd" --nj $nj --feats_dim ${feats_dim} --config_file "$asr_config" --scale 1.0
 fi
 
-token_list=${feats_dir}/data/${lang}_token_list/char/tokens.txt
+token_list=${feats_dir}/data/${lang}_token_list/$token_type/tokens.txt
 echo "dictionary: ${token_list}"
 if [ ${stage} -le 2 ] && [ ${stop_stage} -ge 2 ]; then
     echo "stage 2: Dictionary Preparation"
-    mkdir -p ${feats_dir}/data/${lang}_token_list/char/
+    mkdir -p ${feats_dir}/data/${lang}_token_list/$token_type/
 
     echo "make a dictionary"
     echo "<blank>" > ${token_list}
@@ -134,7 +134,7 @@ if [ ${stage} -le 4 ] && [ ${stop_stage} -ge 4 ]; then
                 --task_name asr \
                 --gpu_id $gpu_id \
                 --use_preprocessor true \
-                --token_type char \
+                --token_type $token_type \
                 --token_list $token_list \
                 --data_dir ${feats_dir}/data \
                 --train_set ${train_set} \

egs/aishell/data2vec_transformer_finetune/run.sh

@@ -91,11 +91,11 @@ if [ ${stage} -le 1 ] && [ ${stop_stage} -ge 1 ]; then
     utils/compute_cmvn.sh --fbankdir ${feats_dir}/data/${train_set} --cmd "$train_cmd" --nj $nj --feats_dim ${feats_dim} --config_file "$asr_config" --scale 1.0
 fi
 
-token_list=${feats_dir}/data/${lang}_token_list/char/tokens.txt
+token_list=${feats_dir}/data/${lang}_token_list/$token_type/tokens.txt
 echo "dictionary: ${token_list}"
 if [ ${stage} -le 2 ] && [ ${stop_stage} -ge 2 ]; then
     echo "stage 2: Dictionary Preparation"
-    mkdir -p ${feats_dir}/data/${lang}_token_list/char/
+    mkdir -p ${feats_dir}/data/${lang}_token_list/$token_type/
 
     echo "make a dictionary"
     echo "<blank>" > ${token_list}
@@ -134,7 +134,7 @@ if [ ${stage} -le 4 ] && [ ${stop_stage} -ge 4 ]; then
                 --task_name asr \
                 --gpu_id $gpu_id \
                 --use_preprocessor true \
-                --token_type char \
+                --token_type $token_type \
                 --token_list $token_list \
                 --data_dir ${feats_dir}/data \
                 --train_set ${train_set} \

egs/aishell/paraformer/run.sh

@@ -88,11 +88,11 @@ if [ ${stage} -le 1 ] && [ ${stop_stage} -ge 1 ]; then
     utils/compute_cmvn.sh --fbankdir ${feats_dir}/data/${train_set} --cmd "$train_cmd" --nj $nj --feats_dim ${feats_dim} --config_file "$asr_config" --scale 1.0
 fi
 
-token_list=${feats_dir}/data/${lang}_token_list/char/tokens.txt
+token_list=${feats_dir}/data/${lang}_token_list/$token_type/tokens.txt
 echo "dictionary: ${token_list}"
 if [ ${stage} -le 2 ] && [ ${stop_stage} -ge 2 ]; then
     echo "stage 2: Dictionary Preparation"
-    mkdir -p ${feats_dir}/data/${lang}_token_list/char/
+    mkdir -p ${feats_dir}/data/${lang}_token_list/$token_type/
    
     echo "make a dictionary"
     echo "<blank>" > ${token_list}
@@ -130,7 +130,7 @@ if [ ${stage} -le 4 ] && [ ${stop_stage} -ge 4 ]; then
                 --task_name asr \
                 --gpu_id $gpu_id \
                 --use_preprocessor true \
-                --token_type char \
+                --token_type $token_type \
                 --token_list $token_list \
                 --dataset_type small \
                 --data_dir ${feats_dir}/data \

egs/aishell/paraformerbert/run.sh

@@ -92,11 +92,11 @@ if [ ${stage} -le 1 ] && [ ${stop_stage} -ge 1 ]; then
     utils/compute_cmvn.sh --fbankdir ${feats_dir}/data/${train_set} --cmd "$train_cmd" --nj $nj --feats_dim ${feats_dim} --config_file "$asr_config" --scale 1.0
 fi
 
-token_list=${feats_dir}/data/${lang}_token_list/char/tokens.txt
+token_list=${feats_dir}/data/${lang}_token_list/$token_type/tokens.txt
 echo "dictionary: ${token_list}"
 if [ ${stage} -le 2 ] && [ ${stop_stage} -ge 2 ]; then
     echo "stage 2: Dictionary Preparation"
-    mkdir -p ${feats_dir}/data/${lang}_token_list/char/
+    mkdir -p ${feats_dir}/data/${lang}_token_list/$token_type/
    
     echo "make a dictionary"
     echo "<blank>" > ${token_list}
@@ -141,7 +141,7 @@ if [ ${stage} -le 4 ] && [ ${stop_stage} -ge 4 ]; then
                 --task_name asr \
                 --gpu_id $gpu_id \
                 --use_preprocessor true \
-                --token_type char \
+                --token_type $token_type \
                 --token_list $token_list \
                 --data_dir ${feats_dir}/data \
                 --train_set ${train_set} \

egs/aishell/transformer/run.sh

@@ -88,11 +88,11 @@ if [ ${stage} -le 1 ] && [ ${stop_stage} -ge 1 ]; then
     utils/compute_cmvn.sh --fbankdir ${feats_dir}/data/${train_set} --cmd "$train_cmd" --nj $nj --feats_dim ${feats_dim} --config_file "$asr_config" --scale 1.0
 fi
 
-token_list=${feats_dir}/data/${lang}_token_list/char/tokens.txt
+token_list=${feats_dir}/data/${lang}_token_list/$token_type/tokens.txt
 echo "dictionary: ${token_list}"
 if [ ${stage} -le 2 ] && [ ${stop_stage} -ge 2 ]; then
     echo "stage 2: Dictionary Preparation"
-    mkdir -p ${feats_dir}/data/${lang}_token_list/char/
+    mkdir -p ${feats_dir}/data/${lang}_token_list/$token_type/
 
     echo "make a dictionary"
     echo "<blank>" > ${token_list}
@@ -130,7 +130,7 @@ if [ ${stage} -le 4 ] && [ ${stop_stage} -ge 4 ]; then
                 --task_name asr \
                 --gpu_id $gpu_id \
                 --use_preprocessor true \
-                --token_type char \
+                --token_type $token_type \
                 --token_list $token_list \
                 --data_dir ${feats_dir}/data \
                 --train_set ${train_set} \

egs/wenetspeech/conformer/run.sh

@@ -84,3 +84,18 @@ if [ ${stage} -le 1 ] && [ ${stop_stage} -ge 1 ]; then
     echo "stage 1: Feature and CMVN Generation"
     utils/compute_cmvn.sh --fbankdir ${feats_dir}/data/${train_set} --cmd "$train_cmd" --nj $nj --feats_dim ${feats_dim} --config_file "$asr_config" --scale 0.1
 fi
+
+token_list=${feats_dir}/data/${lang}_token_list/$token_type/tokens.txt
+echo "dictionary: ${token_list}"
+if [ ${stage} -le 2 ] && [ ${stop_stage} -ge 2 ]; then
+    echo "stage 2: Dictionary Preparation"
+    mkdir -p ${feats_dir}/data/${lang}_token_list/$token_type/
+
+    echo "make a dictionary"
+    echo "<blank>" > ${token_list}
+    echo "<s>" >> ${token_list}
+    echo "</s>" >> ${token_list}
+    utils/text2token.py -s 1 -n 1 --space "" ${feats_dir}/data/$train_set/text | cut -f 2- -d" " | tr " " "\n" \
+        | sort | uniq | grep -a -v -e '^\s*$' | awk '{print $0}' >> ${token_list}
+    echo "<unk>" >> ${token_list}
+fi