diff --git a/examples/aishell/asr0/local/train.sh b/examples/aishell/asr0/local/train.sh
index 7c0ad075..102c051c 100755
--- a/examples/aishell/asr0/local/train.sh
+++ b/examples/aishell/asr0/local/train.sh
@@ -29,6 +29,7 @@ python3 -u ${BIN_DIR}/train.py \
 --seed ${seed}
 else
 python3 -m paddle.distributed.launch --gpus=${CUDA_VISIBLE_DEVICES} ${BIN_DIR}/train.py \
+--ngpu ${ngpu} \
 --config ${config_path} \
 --output exp/${ckpt_name} \
 --model_type ${model_type} \
diff --git a/examples/aishell/asr0/run.sh b/examples/aishell/asr0/run.sh
index 0542e361..114af5a9 100755
--- a/examples/aishell/asr0/run.sh
+++ b/examples/aishell/asr0/run.sh
@@ -2,7 +2,7 @@
 set -e
 source path.sh
 
-gpus=1,2,3
+gpus=0,1,2,3
 stage=0
 stop_stage=100
 conf_path=conf/deepspeech2.yaml    #conf/deepspeech2.yaml or conf/deepspeech2_online.yaml
diff --git a/examples/aishell/asr1/local/train.sh b/examples/aishell/asr1/local/train.sh
index eb9f4f69..5617f7ef 100755
--- a/examples/aishell/asr1/local/train.sh
+++ b/examples/aishell/asr1/local/train.sh
@@ -38,6 +38,7 @@ python3 -u ${BIN_DIR}/train.py \
 --benchmark-max-step ${benchmark_max_step}
 else
 python3 -m paddle.distributed.launch --gpus=${CUDA_VISIBLE_DEVICES} ${BIN_DIR}/train.py \
+--ngpu ${ngpu} \
 --seed ${seed} \
 --config ${config_path} \
 --output exp/${ckpt_name} \
diff --git a/examples/callcenter/asr1/local/train.sh b/examples/callcenter/asr1/local/train.sh
index 3d7d35f1..03b4588e 100755
--- a/examples/callcenter/asr1/local/train.sh
+++ b/examples/callcenter/asr1/local/train.sh
@@ -29,6 +29,7 @@ python3 -u ${BIN_DIR}/train.py \
 --seed ${seed}
 else
 python3 -m paddle.distributed.launch --gpus=${CUDA_VISIBLE_DEVICES} ${BIN_DIR}/train.py \
+--ngpu ${ngpu} \
 --config ${config_path} \
 --output exp/${ckpt_name} \
 --seed ${seed}
diff --git a/examples/librispeech/asr0/local/train.sh b/examples/librispeech/asr0/local/train.sh
index 788f9bf8..50d1d192 100755
--- a/examples/librispeech/asr0/local/train.sh
+++ b/examples/librispeech/asr0/local/train.sh
@@ -29,6 +29,7 @@ python3 -u ${BIN_DIR}/train.py \
 --seed ${seed}
 else
 python3 -m paddle.distributed.launch --gpus=${CUDA_VISIBLE_DEVICES} ${BIN_DIR}/train.py \
+--ngpu ${ngpu} \
 --config ${config_path} \
 --output exp/${ckpt_name} \
 --model_type ${model_type} \
diff --git a/examples/librispeech/asr1/local/train.sh b/examples/librispeech/asr1/local/train.sh
index b21c1017..3860d85c 100755
--- a/examples/librispeech/asr1/local/train.sh
+++ b/examples/librispeech/asr1/local/train.sh
@@ -30,6 +30,7 @@ python3 -u ${BIN_DIR}/train.py \
 --seed ${seed}
 else
 python3 -m paddle.distributed.launch --gpus=${CUDA_VISIBLE_DEVICES} ${BIN_DIR}/train.py \
+--ngpu ${ngpu} \
 --config ${config_path} \
 --output exp/${ckpt_name} \
 --seed ${seed}
diff --git a/examples/librispeech/asr2/local/train.sh b/examples/librispeech/asr2/local/train.sh
index a3ad5fd7..560424ea 100755
--- a/examples/librispeech/asr2/local/train.sh
+++ b/examples/librispeech/asr2/local/train.sh
@@ -28,6 +28,7 @@ python3 -u ${BIN_DIR}/train.py \
 --seed ${seed}
 else
 python3 -m paddle.distributed.launch --gpus=${CUDA_VISIBLE_DEVICES} ${BIN_DIR}/train.py \
+--ngpu ${ngpu} \
 --model-name u2_kaldi \
 --config ${config_path} \
 --output exp/${ckpt_name} \
diff --git a/examples/ted_en_zh/st0/local/train.sh b/examples/ted_en_zh/st0/local/train.sh
index e366376b..ad00653b 100755
--- a/examples/ted_en_zh/st0/local/train.sh
+++ b/examples/ted_en_zh/st0/local/train.sh
@@ -19,11 +19,19 @@ if [ ${seed} != 0 ]; then
     export FLAGS_cudnn_deterministic=True
 fi
 
+if [ ${ngpu} == 0 ]; then
 python3 -u ${BIN_DIR}/train.py \
 --ngpu ${ngpu} \
 --config ${config_path} \
 --output exp/${ckpt_name} \
 --seed ${seed}
+else
+python3 -m paddle.distributed.launch --gpus=${CUDA_VISIBLE_DEVICES} ${BIN_DIR}/train.py \
+--ngpu ${ngpu} \
+--config ${config_path} \
+--output exp/${ckpt_name} \
+--seed ${seed}
+fi
 
 if [ ${seed} != 0 ]; then
     unset FLAGS_cudnn_deterministic
diff --git a/examples/ted_en_zh/st1/local/train.sh b/examples/ted_en_zh/st1/local/train.sh
index a8e4acaa..5da64e99 100755
--- a/examples/ted_en_zh/st1/local/train.sh
+++ b/examples/ted_en_zh/st1/local/train.sh
@@ -20,12 +20,21 @@ if [ ${seed} != 0 ]; then
     export FLAGS_cudnn_deterministic=True
 fi
 
+if [ ${ngpu} == 0 ]; then
 python3 -u ${BIN_DIR}/train.py \
 --ngpu ${ngpu} \
 --config ${config_path} \
 --output exp/${ckpt_name} \
 --checkpoint_path "${ckpt_path}" \
 --seed ${seed}
+else
+python3 -m paddle.distributed.launch --gpus=${CUDA_VISIBLE_DEVICES} ${BIN_DIR}/train.py \
+--ngpu ${ngpu} \
+--config ${config_path} \
+--output exp/${ckpt_name} \
+--checkpoint_path "${ckpt_path}" \
+--seed ${seed}
+fi
 
 if [ ${seed} != 0 ]; then
     unset FLAGS_cudnn_deterministic
@@ -36,4 +45,4 @@ if [ $? -ne 0 ]; then
     exit 1
 fi
 
-exit 0
\ No newline at end of file
+exit 0
diff --git a/examples/timit/asr1/local/train.sh b/examples/timit/asr1/local/train.sh
index c4648c3e..66140758 100755
--- a/examples/timit/asr1/local/train.sh
+++ b/examples/timit/asr1/local/train.sh
@@ -19,10 +19,19 @@ if [ ${seed} != 0  ]; then
     export FLAGS_cudnn_deterministic=True
 fi
 
+if [ ${ngpu} == 0 ]; then
+python3 -u ${BIN_DIR}/train.py \
+--ngpu ${ngpu} \
+--config ${config_path} \
+--output exp/${ckpt_name} \
+--seed ${seed}
+else
 python3 -m paddle.distributed.launch --gpus=${CUDA_VISIBLE_DEVICES} ${BIN_DIR}/train.py \
+--ngpu ${ngpu} \
 --config ${config_path} \
 --output exp/${ckpt_name} \
 --seed ${seed}
+fi
 
 if [ ${seed} != 0 ]; then
     unset FLAGS_cudnn_deterministic
diff --git a/examples/tiny/asr0/local/train.sh b/examples/tiny/asr0/local/train.sh
index 5fecf3de..9060be67 100755
--- a/examples/tiny/asr0/local/train.sh
+++ b/examples/tiny/asr0/local/train.sh
@@ -36,6 +36,7 @@ python3 -u ${BIN_DIR}/train.py \
 --seed ${seed}
 else
 python3 -m paddle.distributed.launch --gpus=${CUDA_VISIBLE_DEVICES} ${BIN_DIR}/train.py \
+--ngpu ${ngpu} \
 --config ${config_path} \
 --output exp/${ckpt_name} \
 --model_type ${model_type} \
diff --git a/examples/tiny/asr1/local/train.sh b/examples/tiny/asr1/local/train.sh
index eb9f4f69..5617f7ef 100755
--- a/examples/tiny/asr1/local/train.sh
+++ b/examples/tiny/asr1/local/train.sh
@@ -38,6 +38,7 @@ python3 -u ${BIN_DIR}/train.py \
 --benchmark-max-step ${benchmark_max_step}
 else
 python3 -m paddle.distributed.launch --gpus=${CUDA_VISIBLE_DEVICES} ${BIN_DIR}/train.py \
+--ngpu ${ngpu} \
 --seed ${seed} \
 --config ${config_path} \
 --output exp/${ckpt_name} \