diff --git a/egs_modelscope/vad/TEMPLATE/README.md b/egs_modelscope/vad/TEMPLATE/README.md
index 4c6f8c282..945f9fa60 100644
--- a/egs_modelscope/vad/TEMPLATE/README.md
+++ b/egs_modelscope/vad/TEMPLATE/README.md
@@ -83,7 +83,7 @@ FunASR also offer recipes [egs_modelscope/vad/TEMPLATE/infer.sh](https://github.
 #### Decode with multi GPUs:
 ```shell
     bash infer.sh \
-    --model "damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch" \
+    --model "damo/speech_fsmn_vad_zh-cn-16k-common-pytorch" \
     --data_dir "./data/test" \
     --output_dir "./results" \
     --batch_size 1 \
@@ -93,7 +93,7 @@ FunASR also offer recipes [egs_modelscope/vad/TEMPLATE/infer.sh](https://github.
 #### Decode with multi-thread CPUs:
 ```shell
     bash infer.sh \
-    --model "damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch" \
+    --model "damo/speech_fsmn_vad_zh-cn-16k-common-pytorch" \
     --data_dir "./data/test" \
     --output_dir "./results" \
     --gpu_inference false \
diff --git a/egs_modelscope/vad/TEMPLATE/infer.py b/egs_modelscope/vad/TEMPLATE/infer.py
index 3d9ee5520..f49ab4b79 100644
--- a/egs_modelscope/vad/TEMPLATE/infer.py
+++ b/egs_modelscope/vad/TEMPLATE/infer.py
@@ -16,10 +16,10 @@ def modelscope_infer(args):
 
 if __name__ == "__main__":
     parser = argparse.ArgumentParser()
-    parser.add_argument('--model', type=str, default="damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch")
+    parser.add_argument('--model', type=str, default="damo/speech_fsmn_vad_zh-cn-16k-common-pytorch")
     parser.add_argument('--audio_in', type=str, default="./data/test/wav.scp")
     parser.add_argument('--output_dir', type=str, default="./results/")
-    parser.add_argument('--batch_size', type=int, default=64)
+    parser.add_argument('--batch_size', type=int, default=1)
     parser.add_argument('--gpuid', type=str, default="0")
     args = parser.parse_args()
     modelscope_infer(args)
\ No newline at end of file
diff --git a/egs_modelscope/vad/speech_fsmn_vad_zh-cn-16k-common/README.md b/egs_modelscope/vad/speech_fsmn_vad_zh-cn-16k-common/README.md
index bb55ab52e..92088a21d 120000
--- a/egs_modelscope/vad/speech_fsmn_vad_zh-cn-16k-common/README.md
+++ b/egs_modelscope/vad/speech_fsmn_vad_zh-cn-16k-common/README.md
@@ -1 +1 @@
-../../TEMPLATE/README.md
\ No newline at end of file
+../TEMPLATE/README.md
\ No newline at end of file
diff --git a/egs_modelscope/vad/speech_fsmn_vad_zh-cn-16k-common/demo.py b/egs_modelscope/vad/speech_fsmn_vad_zh-cn-16k-common/demo.py
index bbc16c5b6..eded5edca 100644
--- a/egs_modelscope/vad/speech_fsmn_vad_zh-cn-16k-common/demo.py
+++ b/egs_modelscope/vad/speech_fsmn_vad_zh-cn-16k-common/demo.py
@@ -7,7 +7,7 @@ if __name__ == '__main__':
     inference_pipeline = pipeline(
         task=Tasks.voice_activity_detection,
         model="damo/speech_fsmn_vad_zh-cn-16k-common-pytorch",
-        model_revision='v1.2.0',
+        model_revision=None,
         output_dir=output_dir,
         batch_size=1,
     )
diff --git a/egs_modelscope/vad/speech_fsmn_vad_zh-cn-16k-common/infer.py b/egs_modelscope/vad/speech_fsmn_vad_zh-cn-16k-common/infer.py
index 128fc31c2..f05fbbb8b 120000
--- a/egs_modelscope/vad/speech_fsmn_vad_zh-cn-16k-common/infer.py
+++ b/egs_modelscope/vad/speech_fsmn_vad_zh-cn-16k-common/infer.py
@@ -1 +1 @@
-../../TEMPLATE/infer.py
\ No newline at end of file
+../TEMPLATE/infer.py
\ No newline at end of file
diff --git a/egs_modelscope/vad/speech_fsmn_vad_zh-cn-16k-common/infer.sh b/egs_modelscope/vad/speech_fsmn_vad_zh-cn-16k-common/infer.sh
index 5e59f1841..0b3b38b6f 120000
--- a/egs_modelscope/vad/speech_fsmn_vad_zh-cn-16k-common/infer.sh
+++ b/egs_modelscope/vad/speech_fsmn_vad_zh-cn-16k-common/infer.sh
@@ -1 +1 @@
-../../TEMPLATE/infer.sh
\ No newline at end of file
+../TEMPLATE/infer.sh
\ No newline at end of file
diff --git a/egs_modelscope/vad/speech_fsmn_vad_zh-cn-8k-common/README.md b/egs_modelscope/vad/speech_fsmn_vad_zh-cn-8k-common/README.md
index bb55ab52e..92088a21d 120000
--- a/egs_modelscope/vad/speech_fsmn_vad_zh-cn-8k-common/README.md
+++ b/egs_modelscope/vad/speech_fsmn_vad_zh-cn-8k-common/README.md
@@ -1 +1 @@
-../../TEMPLATE/README.md
\ No newline at end of file
+../TEMPLATE/README.md
\ No newline at end of file
diff --git a/egs_modelscope/vad/speech_fsmn_vad_zh-cn-8k-common/demo.py b/egs_modelscope/vad/speech_fsmn_vad_zh-cn-8k-common/demo.py
index 84863d082..33be5059a 100644
--- a/egs_modelscope/vad/speech_fsmn_vad_zh-cn-8k-common/demo.py
+++ b/egs_modelscope/vad/speech_fsmn_vad_zh-cn-8k-common/demo.py
@@ -7,7 +7,7 @@ if __name__ == '__main__':
     inference_pipeline = pipeline(
         task=Tasks.voice_activity_detection,
         model="damo/speech_fsmn_vad_zh-cn-8k-common",
-        model_revision='v1.2.0',
+        model_revision=None,
         output_dir=output_dir,
         batch_size=1,
     )
diff --git a/egs_modelscope/vad/speech_fsmn_vad_zh-cn-8k-common/demo_online.py b/egs_modelscope/vad/speech_fsmn_vad_zh-cn-8k-common/demo_online.py
index 5b67da74a..ec5c50281 100644
--- a/egs_modelscope/vad/speech_fsmn_vad_zh-cn-8k-common/demo_online.py
+++ b/egs_modelscope/vad/speech_fsmn_vad_zh-cn-8k-common/demo_online.py
@@ -11,7 +11,7 @@ if __name__ == '__main__':
     inference_pipeline = pipeline(
         task=Tasks.voice_activity_detection,
         model="damo/speech_fsmn_vad_zh-cn-8k-common",
-        model_revision='v1.2.0',
+        model_revision=None,
         output_dir=output_dir,
         batch_size=1,
         mode='online',
diff --git a/egs_modelscope/vad/speech_fsmn_vad_zh-cn-8k-common/infer.py b/egs_modelscope/vad/speech_fsmn_vad_zh-cn-8k-common/infer.py
index 128fc31c2..f05fbbb8b 120000
--- a/egs_modelscope/vad/speech_fsmn_vad_zh-cn-8k-common/infer.py
+++ b/egs_modelscope/vad/speech_fsmn_vad_zh-cn-8k-common/infer.py
@@ -1 +1 @@
-../../TEMPLATE/infer.py
\ No newline at end of file
+../TEMPLATE/infer.py
\ No newline at end of file
diff --git a/egs_modelscope/vad/speech_fsmn_vad_zh-cn-8k-common/infer.sh b/egs_modelscope/vad/speech_fsmn_vad_zh-cn-8k-common/infer.sh
index 5e59f1841..0b3b38b6f 120000
--- a/egs_modelscope/vad/speech_fsmn_vad_zh-cn-8k-common/infer.sh
+++ b/egs_modelscope/vad/speech_fsmn_vad_zh-cn-8k-common/infer.sh
@@ -1 +1 @@
-../../TEMPLATE/infer.sh
\ No newline at end of file
+../TEMPLATE/infer.sh
\ No newline at end of file
diff --git a/k2/__init__.py b/k2/__init__.py
new file mode 100644
index 000000000..e69de29bb