Update app.py

app.py

@@ -123,7 +123,7 @@ def speech_generation_fn(story_data):
     story_gen_agent = MMStoryAgent()
     story_gen_agent.generate_speech(config, story_data)
 
-@spaces.GPU(duration=120)
+@spaces.GPU(duration=60)
 def sound_generation_fn(sound_guidance_scale, sound_seed, n_candidate_per_text,
                         story_data, progress=gr.Progress(track_tqdm=True)):
     deep_update(config, {

mm_story_agent/__init__.py

@@ -36,21 +36,25 @@ class MMStoryAgent:
     
     def generate_speech(self, config, pages):
         story_dir = Path(config["story_dir"])
+        (story_dir / "speech").mkdir(exist_ok=True, parents=True)
         speech_agent = CosyVoiceAgent(config["speech_generation"])
         speech_agent.call(pages, story_dir / "speech")
     
     def generate_sound(self, config, pages):
         story_dir = Path(config["story_dir"])
+        (story_dir / "sound").mkdir(exist_ok=True, parents=True)
         sound_agent = AudioLDM2Agent(config["sound_generation"])
         sound_agent.call(pages, story_dir / "sound")
 
     def generate_music(self, config, pages):
         story_dir = Path(config["story_dir"])
+        (story_dir / "music").mkdir(exist_ok=True, parents=True)
         music_agent = MusicGenAgent(config["music_generation"])
         music_agent.call(pages, story_dir / "music")
     
     def generate_image(self, config, pages):
         story_dir = Path(config["story_dir"])
+        (story_dir / "image").mkdir(exist_ok=True, parents=True)
         image_agent = StoryDiffusionAgent(config["image_generation"])
         image_agent.call(pages, story_dir / "image")