Update app.py

app.py

@@ -10,7 +10,6 @@ from gtts import gTTS
 from PIL import Image
 import numpy as np
 from nltk.tokenize import sent_tokenize
-from IPython.display import Audio
 import spaces
 
 device = "cuda" if torch.cuda.is_available() else "cpu"
@@ -22,22 +21,21 @@ if not api_key:
 client = OpenAI(api_key=api_key)
 
 yolo_model = YOLO("yolov8s.pt")
+
 stable_diffusion = StableDiffusionPipeline.from_pretrained(
     "runwayml/stable-diffusion-v1-5",
-    torch_dtype=torch.float16,
-    safety_checker=None
+    torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32
 ).to(device)
-stable_diffusion.to(device)
+
 nltk.download("punkt")
 
 summarizer = pipeline(
     "summarization", 
     model="sshleifer/distilbart-cnn-6-6"
 )
-
 @spaces.GPU
 def detect_objects(image):
-    image_array = np.array(image)
+    image_array = np.array(image)  # Μετατροπή PIL → NumPy
     results = yolo_model(image_array)
     detected_objects = []
     for r in results:
@@ -49,12 +47,12 @@ def detect_objects(image):
 
 def generate_story(detected_objects):
     story_prompt = f"Write a short story based on the following objects: {', '.join(detected_objects)}"
-    response = client.completions.create(
+    response = client.chat.completions.create(
         model="gpt-4o-mini",
         messages=[{"role": "user", "content": story_prompt}],
         max_tokens=200
     )
-    return response.choices[0].text.strip()  # Διορθώθηκε
+    return response.choices[0].message.content.strip()
 
 def summarize_story(story):
     summary = summarizer(story, max_length=100, do_sample=False)[0]['summary_text']
@@ -66,7 +64,7 @@ def generate_images(story):
     prompts = [f"Highly detailed, cinematic scene: {scene}, digital art, 4K, realistic lighting" for scene in scenes]
     images = []
     for prompt in prompts:
-        image = stable_diffusion(prompt=prompt).images[0]  # Διορθώθηκε
+        image = stable_diffusion(prompt).images[0]  # Διόρθωση
         images.append(image)
     return images
 
@@ -77,9 +75,7 @@ def text_to_speech(story):
     return audio_file_path
 
 def full_pipeline(image):
-    image_path = "input.jpg"
-    image.save(image_path)  # Διορθώθηκε
-    detected_objects = detect_objects(image_path)
+    detected_objects = detect_objects(image)
     story = generate_story(detected_objects)
     scenes = summarize_story(story)
     images = generate_images(story)
@@ -103,3 +99,4 @@ demo = gr.Interface(
 
 if __name__ == "__main__":
     demo.launch()
+