Update memo/utils/vision_utils.py

memo/utils/vision_utils.py

@@ -35,8 +35,8 @@ def tensor_to_video(tensor, output_video_path, input_audio_path, fps=30):
     final_duration = min(video_duration, audio_duration)
     audio_clip = audio_clip.subclip(0, final_duration)
     new_video_clip = VideoClip(make_frame, duration=final_duration)
-    new_video_clip = new_video_clip.set_audio(audio_clip), codec="libx264", audio_codec="aac")
-
+    new_video_clip = new_video_clip.set_audio(audio_clip)
+    new_video_clip.write_videofile(output_video_path, fps=fps, codec="libx264", audio_codec="aac")
 
 @torch.no_grad()
 def preprocess_image(face_analysis_model: str, image_path: str, image_size: int = 512):