Update app.py

app.py

@@ -9,6 +9,7 @@ from diffusers import DiffusionPipeline
 import google.generativeai as genai
 from pathlib import Path
 
+
 # Load environment variables from .env file
 load_dotenv()
 
@@ -23,12 +24,12 @@ def analyze_image_with_gemini(image_file):
     Analyzes an uploaded image with Gemini and generates a descriptive caption.
     """
     try:
-        # Save uploaded image to a temporary file
+# Save uploaded image to a temporary file
         temp_image_path = tempfile.NamedTemporaryFile(delete=False, suffix=".jpg").name
         with open(temp_image_path, "wb") as temp_file:
             temp_file.write(image_file)
 
-        # Prepare the image data and prompt for Gemini
+# Prepare the image data and prompt for Gemini
         image_parts = [{"mime_type": "image/jpeg", "data": Path(temp_image_path).read_bytes()}]
         prompt_parts = ["Describe precisely the image in one sentence.\n", image_parts[0], "\n"]
         generation_config = {"temperature": 0.05, "top_p": 1, "top_k": 26, "max_output_tokens": 4096}
@@ -49,11 +50,11 @@ def get_audioldm_from_caption(caption):
     """
     Generates sound from a caption using the AudioLDM-2 model.
     """
-    # Initialize the model
+# Initialize the model
     pipe = DiffusionPipeline.from_pretrained("cvssp/audioldm2", use_auth_token=hf_token)
     pipe = pipe.to("cuda" if torch.cuda.is_available() else "cpu")
 
-    # Generate audio from the caption
+# Generate audio from the caption
     audio_output = pipe(prompt=caption, num_inference_steps=50, guidance_scale=7.5)
     audio = audio_output.audios[0]  
 
@@ -67,22 +68,22 @@ css="""
 #col-container{
     margin: 0 auto;
     max-width: 800px;
-}
+    }
+
 """
 
 # Gradio interface setup
-@spaces.GPU
 with gr.Blocks(css=css) as demo:
     # Main Title and App Description
     with gr.Column(elem_id="col-container"):
         gr.HTML("""
-        <h1 style="text-align: center;">
-            🎶 Generate Sound Effects from Image
+    <h1 style="text-align: center;">
+           🎶 Generate Sound Effects from Image
         </h1>
-        <p style="text-align: center;">
-            ⚡ Powered by <a href="https://bilsimaging.com" _blank >Bilsimaging</a>
+         <p style="text-align: center;">
+          ⚡  Powered by <a href="https://bilsimaging.com" _blank >Bilsimaging</a>
         </p>
-        """)
+    """)
     
     gr.Markdown("""
     Welcome to this unique sound effect generator! This tool allows you to upload an image and generate a descriptive caption and a corresponding sound effect. Whether you're exploring the sound of nature, urban environments, or anything in between, this app brings your images to auditory life.
@@ -94,28 +95,26 @@ with gr.Blocks(css=css) as demo:
     
     Enjoy the journey from visual to auditory sensation with just a few clicks!
     
-    For example demo sound effects generated, check out our [YouTube channel](https://www.youtube.com/playlist?list=PLwEbW4bdYBSC8exiJ9PfzufGND_14f--C)
+    For Example Demos sound effects generated , check out our [YouTube channel](https://www.youtube.com/playlist?list=PLwEbW4bdYBSC8exiJ9PfzufGND_14f--C)
     """)
     
-    # Interface Components
+# Interface Components
     image_upload = gr.File(label="Upload Image", type="binary")
     generate_description_button = gr.Button("Tap to Generate a Description from your image")
     caption_display = gr.Textbox(label="Image Description", interactive=False)  # Keep as read-only
     generate_sound_button = gr.Button("Generate Sound Effect")
     audio_output = gr.Audio(label="Generated Sound Effect")
-
-    # Extra footer
+# extra footer
     gr.Markdown("""## 👥 How You Can Contribute
         We welcome contributions and suggestions for improvements. Your feedback is invaluable to the continuous enhancement of this application. 
                 
         For support, questions, or to contribute, please contact us at [[email protected]](mailto:[email protected]).
                 
         Support our work and get involved by donating through [Ko-fi](https://ko-fi.com/bilsimaging). - Bilel Aroua
-    """)
+            """)
     gr.Markdown("""## 📢 Stay Connected
-        This app is a testament to the creative possibilities that emerge when technology meets art. Enjoy exploring the auditory landscape of your images!
-    """)
-    
+        this app is a testament to the creative possibilities that emerge when technology meets art. Enjoy exploring the auditory landscape of your images!
+            """)
     # Function to update the caption display based on the uploaded image
     def update_caption(image_file):
         description, _ = analyze_image_with_gemini(image_file)
@@ -138,5 +137,7 @@ with gr.Blocks(css=css) as demo:
         outputs=audio_output
     )
 
+
+
 # Launch the Gradio app
 demo.launch(debug=True, share=True)