text to video

app.py

@@ -95,14 +95,23 @@ class Client:
         self.add_message(message, "assistant")
         messages[-1][1] = message
         for image_url in image_urls:
-            image_url = image_url.replace("public/", "")
-            messages = messages + [((None, (f"public/{image_url}",)))]
+            if not image_url.startswith("http"):
+                image_url = image_url.replace("public/", "")
+                messages = messages + [((None, (f"public/{image_url}",)))]
+            else:
+                messages = messages + [((None, (f"{image_url}",)))]
         for audio_url in audio_urls:
-            audio_url = audio_url.replace("public/", "")
-            messages = messages + [((None, (f"public/{audio_url}",)))]
+            if not audio_url.startswith("http"):
+                audio_url = audio_url.replace("public/", "")
+                messages = messages + [((None, (f"public/{audio_url}",)))]
+            else:
+                messages = messages + [((None, (f"{audio_url}",)))]
         for video_url in video_urls:
-            video_url = video_url.replace("public/", "")
-            messages = messages + [((None, (f"public/{video_url}",)))]
+            if not video_url.startswith("http"):
+                video_url = video_url.replace("public/", "")
+                messages = messages + [((None, (f"public/{video_url}",)))]
+            else:
+                messages = messages + [((None, (f"{video_url}",)))]
         return messages
 
 with gr.Blocks() as demo:
@@ -168,7 +177,6 @@ with gr.Blocks() as demo:
                 "Please generate a canny image based on /examples/f.jpg",
                 "show me a joke and an image of cat",
                 "what is in the examples/a.jpg",
-                "generate a video and audio about a dog is running on the grass",
                 "based on the /examples/a.jpg, please generate a video and audio",
                 "based on pose of /examples/d.jpg and content of /examples/e.jpg, please show me a new image",
                 ],

awesome_chat.py

@@ -835,12 +835,14 @@ def chat_huggingface(messages, openaikey = None, huggingfacetoken = None, return
     logger.info("*"*80)
     logger.info(f"input: {input}")
 
-    task_str = parse_task(context, input, openaikey).strip()
+    task_str = parse_task(context, input, openaikey)
     logger.info(task_str)
 
     if "error" in task_str:
         return {"message": "You exceeded your current quota, please check your plan and billing details."}
-
+    else:
+        task_str = task_str.strip()
+        
     if task_str == "[]":  # using LLM response for empty task
         record_case(success=False, **{"input": input, "task": [], "reason": "task parsing fail: empty", "op": "chitchat"})
         response = chitchat(messages, openaikey)

models_server.py

@@ -374,10 +374,16 @@ def models(model_id, data):
             # pipe.enable_model_cpu_offload()
             prompt = data["text"]
             video_frames = pipe(prompt, num_inference_steps=50, num_frames=40).frames
-            video_path = export_to_video(video_frames)
             file_name = str(uuid.uuid4())[:4]
-            os.system(f"LD_LIBRARY_PATH=/usr/local/lib /usr/local/bin/ffmpeg -i {video_path} -vcodec libx264 public/videos/{file_name}.mp4")
-            result = {"path": f"/videos/{file_name}.mp4"}
+            video_path = export_to_video(video_frames, f"public/videos/{file_name}.mp4")
+            
+            new_file_name = str(uuid.uuid4())[:4]
+            os.system(f"ffmpeg -i {video_path} -vcodec libx264 public/videos/{new_file_name}.mp4")
+
+            if os.path.exists(f"public/videos/{new_file_name}.mp4"): 
+                result = {"path": f"/videos/{new_file_name}.mp4"}
+            else:
+                result = {"path": f"/videos/{file_name}.mp4"}
 
         # controlnet
         if model_id.startswith("lllyasviel/sd-controlnet-"):