diff --git a/examples/vctk/tts3/local/synthesize.sh b/examples/vctk/tts3/local/synthesize.sh
index 8381af464e62a3e4def2cae1e5844dc7778f853d..9e03f9b8a4701c0232ad7a2f1262533dd0a61435 100755
--- a/examples/vctk/tts3/local/synthesize.sh
+++ b/examples/vctk/tts3/local/synthesize.sh
@@ -4,18 +4,43 @@ config_path=$1
 train_output_path=$2
 ckpt_name=$3
 
-FLAGS_allocator_strategy=naive_best_fit \
-FLAGS_fraction_of_gpu_memory_to_use=0.01 \
-python3 ${BIN_DIR}/../synthesize.py \
-    --am=fastspeech2_vctk \
-    --am_config=${config_path} \
-    --am_ckpt=${train_output_path}/checkpoints/${ckpt_name} \
-    --am_stat=dump/train/speech_stats.npy \
-    --voc=pwgan_vctk \
-    --voc_config=pwg_vctk_ckpt_0.1.1/default.yaml  \
-    --voc_ckpt=pwg_vctk_ckpt_0.1.1/snapshot_iter_1500000.pdz \
-    --voc_stat=pwg_vctk_ckpt_0.1.1/feats_stats.npy \
-    --test_metadata=dump/test/norm/metadata.jsonl \
-    --output_dir=${train_output_path}/test \
-    --phones_dict=dump/phone_id_map.txt \
-    --speaker_dict=dump/speaker_id_map.txt
+stage=0
+stop_stage=0
+
+# pwgan
+if [ ${stage} -le 0 ] && [ ${stop_stage} -ge 0 ]; then
+    FLAGS_allocator_strategy=naive_best_fit \
+    FLAGS_fraction_of_gpu_memory_to_use=0.01 \
+    python3 ${BIN_DIR}/../synthesize.py \
+        --am=fastspeech2_vctk \
+        --am_config=${config_path} \
+        --am_ckpt=${train_output_path}/checkpoints/${ckpt_name} \
+        --am_stat=dump/train/speech_stats.npy \
+        --voc=pwgan_vctk \
+        --voc_config=pwg_vctk_ckpt_0.1.1/default.yaml  \
+        --voc_ckpt=pwg_vctk_ckpt_0.1.1/snapshot_iter_1500000.pdz \
+        --voc_stat=pwg_vctk_ckpt_0.1.1/feats_stats.npy \
+        --test_metadata=dump/test/norm/metadata.jsonl \
+        --output_dir=${train_output_path}/test \
+        --phones_dict=dump/phone_id_map.txt \
+        --speaker_dict=dump/speaker_id_map.txt
+fi
+
+# hifigan
+if [ ${stage} -le 1 ] && [ ${stop_stage} -ge 1 ]; then
+    FLAGS_allocator_strategy=naive_best_fit \
+    FLAGS_fraction_of_gpu_memory_to_use=0.01 \
+    python3 ${BIN_DIR}/../synthesize.py \
+        --am=fastspeech2_aishell3 \
+        --am_config=${config_path} \
+        --am_ckpt=${train_output_path}/checkpoints/${ckpt_name} \
+        --am_stat=dump/train/speech_stats.npy \
+        --voc=hifigan_vctk \
+        --voc_config=hifigan_vctk_ckpt_0.2.0/default.yaml  \
+        --voc_ckpt=hifigan_vctk_ckpt_0.2.0/snapshot_iter_2500000.pdz \
+        --voc_stat=hifigan_vctk_ckpt_0.2.0/feats_stats.npy \
+        --test_metadata=dump/test/norm/metadata.jsonl \
+        --output_dir=${train_output_path}/test \
+        --phones_dict=dump/phone_id_map.txt \
+        --speaker_dict=dump/speaker_id_map.txt
+fi
diff --git a/examples/vctk/tts3/local/synthesize_e2e.sh b/examples/vctk/tts3/local/synthesize_e2e.sh
index 60d56d1c9cb5d09544f4edee84e99618912ea643..16b0d62ba39270d37d952fa0a40ddcf0ad5b79fd 100755
--- a/examples/vctk/tts3/local/synthesize_e2e.sh
+++ b/examples/vctk/tts3/local/synthesize_e2e.sh
@@ -4,21 +4,49 @@ config_path=$1
 train_output_path=$2
 ckpt_name=$3
 
-FLAGS_allocator_strategy=naive_best_fit \
-FLAGS_fraction_of_gpu_memory_to_use=0.01 \
-python3 ${BIN_DIR}/../synthesize_e2e.py \
-    --am=fastspeech2_vctk \
-    --am_config=${config_path} \
-    --am_ckpt=${train_output_path}/checkpoints/${ckpt_name} \
-    --am_stat=dump/train/speech_stats.npy \
-    --voc=pwgan_vctk \
-    --voc_config=pwg_vctk_ckpt_0.1.1/default.yaml  \
-    --voc_ckpt=pwg_vctk_ckpt_0.1.1/snapshot_iter_1500000.pdz \
-    --voc_stat=pwg_vctk_ckpt_0.1.1/feats_stats.npy \
-    --lang=en \
-    --text=${BIN_DIR}/../sentences_en.txt \
-    --output_dir=${train_output_path}/test_e2e \
-    --phones_dict=dump/phone_id_map.txt \
-    --speaker_dict=dump/speaker_id_map.txt \
-    --spk_id=0 \
-    --inference_dir=${train_output_path}/inference
+stage=0
+stop_stage=0
+
+# pwgan
+if [ ${stage} -le 0 ] && [ ${stop_stage} -ge 0 ]; then
+    FLAGS_allocator_strategy=naive_best_fit \
+    FLAGS_fraction_of_gpu_memory_to_use=0.01 \
+    python3 ${BIN_DIR}/../synthesize_e2e.py \
+        --am=fastspeech2_vctk \
+        --am_config=${config_path} \
+        --am_ckpt=${train_output_path}/checkpoints/${ckpt_name} \
+        --am_stat=dump/train/speech_stats.npy \
+        --voc=pwgan_vctk \
+        --voc_config=pwg_vctk_ckpt_0.1.1/default.yaml  \
+        --voc_ckpt=pwg_vctk_ckpt_0.1.1/snapshot_iter_1500000.pdz \
+        --voc_stat=pwg_vctk_ckpt_0.1.1/feats_stats.npy \
+        --lang=en \
+        --text=${BIN_DIR}/../sentences_en.txt \
+        --output_dir=${train_output_path}/test_e2e \
+        --phones_dict=dump/phone_id_map.txt \
+        --speaker_dict=dump/speaker_id_map.txt \
+        --spk_id=0 \
+        --inference_dir=${train_output_path}/inference
+fi
+
+# hifigan
+if [ ${stage} -le 1 ] && [ ${stop_stage} -ge 1 ]; then
+    FLAGS_allocator_strategy=naive_best_fit \
+    FLAGS_fraction_of_gpu_memory_to_use=0.01 \
+    python3 ${BIN_DIR}/../synthesize_e2e.py \
+        --am=fastspeech2_vctk \
+        --am_config=${config_path} \
+        --am_ckpt=${train_output_path}/checkpoints/${ckpt_name} \
+        --am_stat=dump/train/speech_stats.npy \
+        --voc=hifigan_vctk \
+        --voc_config=hifigan_vctk_ckpt_0.2.0/default.yaml  \
+        --voc_ckpt=hifigan_vctk_ckpt_0.2.0/snapshot_iter_2500000.pdz \
+        --voc_stat=hifigan_vctk_ckpt_0.2.0/feats_stats.npy \
+        --lang=en \
+        --text=${BIN_DIR}/../sentences_en.txt \
+        --output_dir=${train_output_path}/test_e2e \
+        --phones_dict=dump/phone_id_map.txt \
+        --speaker_dict=dump/speaker_id_map.txt \
+        --spk_id=0 \
+        --inference_dir=${train_output_path}/inference
+fi
\ No newline at end of file