Update app.py

app.py

@@ -5,7 +5,7 @@ from moviepy.editor import ImageSequenceClip, concatenate_videoclips
 from PIL import Image
 import torch
 from diffusers import AudioLDMPipeline
-from transformers import AutoProcessor, ClapModel
+from transformers import AutoProcessor, ClapModel, BlipProcessor, BlipForConditionalGeneration
 
 # make Space compatible with CPU duplicates
 if torch.cuda.is_available():
@@ -26,6 +26,10 @@ processor = AutoProcessor.from_pretrained("sanchit-gandhi/clap-htsat-unfused-m-f
 
 generator = torch.Generator(device)
 
+# Charger le modèle et le processeur Blip pour la description d'images
+image_caption_processor = BlipProcessor.from_pretrained("Salesforce/blip-image-captioning-base")
+image_caption_model = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-base")
+
 # Streamlit app setup
 st.set_page_config(
     page_title="Text to Media",
@@ -54,9 +58,9 @@ if uploaded_files:
         # Générez la légende pour chaque image
         try:
             image = Image.open(image_path).convert("RGB")
-            inputs = processor(image, return_tensors="pt")
-            out = model.generate(**inputs)
-            caption = processor.decode(out[0], skip_special_tokens=True)
+            inputs = image_caption_processor(image, return_tensors="pt")
+            out = image_caption_model.generate(**inputs)
+            caption = image_caption_processor.decode(out[0], skip_special_tokens=True)
             descriptions.append(caption)
         except Exception as e:
             descriptions.append("Erreur lors de la génération de la légende")