Update app.py

app.py

@@ -103,6 +103,29 @@ def process_speech(input_language, audio_input):
     except Exception as e :
         return f"{e}"
 
+def save_image(image_input, output_dir="saved_images"):
+    if not os.path.exists(output_dir):
+        os.makedirs(output_dir)
+
+    # Generate a unique file name
+    file_name = f"image_{int(time.time())}.png"
+    file_path = os.path.join(output_dir, file_name)
+
+    # Check the type of image_input and handle accordingly
+    if isinstance(image_input, np.ndarray):  # If image_input is a NumPy array
+        Image.fromarray(image_input).save(file_path)
+    elif isinstance(image_input, Image.Image):  # If image_input is a PIL image
+        image_input.save(file_path)
+    elif isinstance(image_input, str) and image_input.startswith('data:image'):  # If image_input is a base64 string
+        image_data = base64.b64decode(image_input.split(',')[1])
+        with open(file_path, 'wb') as f:
+            f.write(image_data)
+    else:
+        raise ValueError("Unsupported image format")
+
+    return file_path
+
+
 def process_image(image_input):
     # Initialize the Gradio client with the URL of the Gradio server
     client = Client("https://adept-fuyu-8b-demo.hf.space/--replicas/pqjvl/")
@@ -304,7 +327,9 @@ def process_and_query(input_language=None, audio_input=None, image_input=None, t
         
         # Process image input
         if image_input is not None:
-            image_text = process_image(image_input)
+            # Convert image_input to a file path
+            image_file_path = save_image(image_input)
+            image_text = process_image(image_file_path)
             combined_text += "\n\n**Image Input:**\n" + image_text
 
         # Process audio input