fix

infer.py

@@ -11,6 +11,8 @@ from model import FantasyTalkingAudioConditionModel
 from utils import save_video, get_audio_features, resize_image_by_longest_edge
 from pathlib import Path
 from datetime import datetime
+from modelscope import snapshot_download
+
 
 def parse_args():
     parser = argparse.ArgumentParser(description="Simple example of a training script.")
@@ -122,14 +124,14 @@ def parse_args():
 
 def load_models(args):
     # Load Wan I2V models
+
+    snapshot_download("Wan-AI/Wan2.1-I2V-14B-720P", local_dir="models/Wan2.1-I2V-14B-720P")
+    snapshot_download("facebook/wav2vec2-base-960h", local_dir="models/wav2vec2-base-960h")
+    snapshot_download("acvlab/FantasyTalking", local_dir="models")
+
+
     model_manager = ModelManager(device="cpu")
-    ckpt_path = hf_hub_download(configs[name].repo_id, configs[name].repo_flow.replace("sft", "safetensors"))
     
-    if hf_download:
-        # lora_ckpt_path = hf_hub_download("bytedance-research/UNO", "dit_lora.safetensors")
-        try:
-            lora_ckpt_path = hf_hub_download("bytedance-research/UNO", "dit_lora.safetensors")
-            
     model_manager.load_models(
         [
             [