example added + colors

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+*.mp3 filter=lfs diff=lfs merge=lfs -text

app.py

@@ -18,6 +18,7 @@ MODEL_IDS = {
 device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
 model_cache = {}
 
+
 def load_model(model_name):
     """Load model if not already cached"""
     if model_name not in model_cache:
@@ -28,109 +29,262 @@ def load_model(model_name):
         model_cache[model_name] = model
     return model_cache[model_name]
 
+
 def process_audio(audio_path, model_name):
     """Process audio file and return prediction"""
     try:
         model = load_model(model_name)
         max_time = model.config.audio.max_time
-        
+
         # Load and process audio
         audio, sr = librosa.load(audio_path, sr=16000)
         chunk_samples = int(max_time * sr)
         total_chunks = len(audio) // chunk_samples
         middle_chunk_idx = total_chunks // 2
-        
+
         # Extract middle chunk
         start = middle_chunk_idx * chunk_samples
         end = start + chunk_samples
         chunk = audio[start:end]
-        
+
         if len(chunk) < chunk_samples:
             chunk = np.pad(chunk, (0, chunk_samples - len(chunk)))
-            
+
         # Get prediction
         with torch.no_grad():
             chunk = torch.from_numpy(chunk).float().to(device)
             pred = model(chunk.unsqueeze(0))
             prob = torch.sigmoid(pred).cpu().numpy()[0]
-            
-        return {"Real": 1 - prob, "Fake": prob}
+
+        real_prob = 1 - prob
+        fake_prob = prob
+        
+        # Return formatted results with emojis
+        return {
+            "🎵 Real": float(real_prob),
+            "🤖 Fake": float(fake_prob)
+        }
 
     except Exception as e:
-        return {"Error": str(e)}
+        return {"❌ Error": str(e)}
+
 
 def predict(audio_file, model_name):
     """Gradio interface function"""
     if audio_file is None:
-        return {"Message": "Please upload an audio file"}
+        return {"⚠️ Message": "Please upload an audio file"}
     return process_audio(audio_file, model_name)
 
+
+# Custom CSS for styling
+css = """
+:root {
+    --primary-color: #6366f1;
+    --secondary-color: #8b5cf6;
+    --accent-color: #ec4899;
+    --background-color: #f8fafc;
+    --text-color: #1e293b;
+    --border-radius: 10px;
+}
+
+.gradio-container {
+    background-color: var(--background-color);
+}
+
+.gr-button {
+    background: linear-gradient(90deg, var(--primary-color), var(--secondary-color));
+    border: none !important;
+    color: white !important;
+    border-radius: var(--border-radius) !important;
+}
+
+.gr-button:hover {
+    background: linear-gradient(90deg, var(--secondary-color), var(--accent-color));
+    transform: translateY(-2px);
+    box-shadow: 0 10px 20px rgba(0,0,0,0.1);
+    transition: all 0.3s ease;
+}
+
+.gr-form {
+    border-radius: var(--border-radius) !important;
+    border: 1px solid #e2e8f0 !important;
+    box-shadow: 0 4px 12px rgba(0,0,0,0.05) !important;
+}
+
+.footer {
+    margin-top: 20px;
+    text-align: center;
+    font-size: 0.9em;
+    color: #64748b;
+}
+
+.gradient-text {
+    background: linear-gradient(90deg, var(--primary-color), var(--accent-color));
+    -webkit-background-clip: text;
+    -webkit-text-fill-color: transparent;
+    background-clip: text;
+    text-fill-color: transparent;
+}
+
+.logo-container {
+    display: flex;
+    justify-content: center;
+    margin-bottom: 1rem;
+}
+
+.header-container {
+    text-align: center;
+    margin-bottom: 2rem;
+    padding: 1.5rem;
+    background: rgba(255, 255, 255, 0.8);
+    border-radius: var(--border-radius);
+    box-shadow: 0 4px 15px rgba(0, 0, 0, 0.05);
+}
+
+.resource-links {
+    display: flex;
+    justify-content: center;
+    gap: 1rem;
+    flex-wrap: wrap;
+    margin-bottom: 1.5rem;
+}
+
+.resource-link {
+    display: inline-block;
+    padding: 0.5rem 1rem;
+    background: white;
+    border-radius: var(--border-radius);
+    color: var(--primary-color);
+    text-decoration: none;
+    box-shadow: 0 2px 5px rgba(0, 0, 0, 0.1);
+    transition: all 0.2s ease;
+}
+
+.resource-link:hover {
+    transform: translateY(-2px);
+    box-shadow: 0 4px 8px rgba(0, 0, 0, 0.15);
+}
+
+.label-container {
+    border-radius: var(--border-radius);
+    overflow: hidden;
+    box-shadow: 0 4px 12px rgba(0,0,0,0.05);
+}
+"""
+
 # Create Gradio interface
-with gr.Blocks() as demo:
+with gr.Blocks(css=css) as demo:
     # Title, Subtitle, and Logo
     gr.HTML(
         """
-        <div style="text-align: center;">
-            <img src="https://i.postimg.cc/3Jx3yZ5b/real-vs-fake-sonics-w-logo.jpg" 
-                 style="max-width: 150px; margin: 0 auto;">
-            <h1>SONICS: Synthetic Or Not - Identifying Counterfeit Songs</h1>
+        <div class="header-container">
+            <div class="logo-container">
+                <img src="https://i.postimg.cc/3Jx3yZ5b/real-vs-fake-sonics-w-logo.jpg" 
+                     style="max-width: 180px; border-radius: 15px; box-shadow: 0 4px 12px rgba(0,0,0,0.1);">
+            </div>
+            <h1 class="gradient-text">🎵 SONICS: Synthetic Or Not - Identifying Counterfeit Songs 🤖</h1>
             <h3>ICLR 2025 [Poster]</h3>
-            <p style="font-size: 1.1em; color: #666; margin: 10px 0;">
-                Detect if a song is real or AI-generated (created using text-to-song models).
-                Upload any audio file to check its authenticity!
+            <p style="font-size: 1.1em; color: #64748b; margin: 15px 0;">
+                Detect if a song is real or AI-generated with our state-of-the-art models.
+                Simply upload an audio file to verify its authenticity!
             </p>
         </div>
         """
     )
-    
-    # # Resource Links
-    # with gr.Row():
-    #     paper_radio = gr.Radio(
-    #         choices=["Paper", "Dataset", "ArXiv", "GitHub"],
-    #         label="Resources",
-    #         info="Click to visit respective links"
-    #     )
-        
+
+    # Resource Links
     gr.HTML(
         """
-        <div style="text-align: center; margin-bottom: 1rem;">
-            <p>
-                <a href="https://openreview.net/forum?id=PY7KSh29Z8" target="_blank">📄 Paper</a> |
-                <a href="https://huggingface.co/datasets/awsaf49/sonics" target="_blank">🎵 Dataset</a> |
-                <a href="https://huggingface.co/collections/awsaf49/sonics-spectttra-67bb6517b3920fd18e409013" target="_blank">🤖 Models</a> |
-                <a href="https://arxiv.org/abs/2408.14080" target="_blank">🔬 ArXiv</a> |
-                <a href="https://github.com/awsaf49/sonics" target="_blank">💻 GitHub</a>
-            </p>
+        <div class="resource-links">
+            <a href="https://openreview.net/forum?id=PY7KSh29Z8" target="_blank" class="resource-link">
+                📄 Paper
+            </a>
+            <a href="https://huggingface.co/datasets/awsaf49/sonics" target="_blank" class="resource-link">
+                🎵 Dataset
+            </a>
+            <a href="https://huggingface.co/collections/awsaf49/sonics-spectttra-67bb6517b3920fd18e409013" target="_blank" class="resource-link">
+                🤖 Models
+            </a>
+            <a href="https://arxiv.org/abs/2408.14080" target="_blank" class="resource-link">
+                🔬 ArXiv
+            </a>
+            <a href="https://github.com/awsaf49/sonics" target="_blank" class="resource-link">
+                💻 GitHub
+            </a>
         </div>
         """
     )
-    
+
     # Main Interface
-    with gr.Row():
+    with gr.Row(equal_height=True):
         with gr.Column():
             audio_input = gr.Audio(
-                label="Upload Audio File",
-                type="filepath"
+                label="🎧 Upload Audio File",
+                type="filepath",
+                elem_id="audio_input"
             )
+            
             model_dropdown = gr.Dropdown(
                 choices=list(MODEL_IDS.keys()),
                 value="SpecTTTra-γ (5s)",
-                label="Select Model"
+                label="🔍 Select Model",
+                elem_id="model_dropdown"
             )
-            submit_btn = gr.Button("Analyze Audio")
-        
+            
+            submit_btn = gr.Button(
+                "✨ Analyze Audio",
+                elem_id="submit_btn"
+            )
+
         with gr.Column():
+            # Define output before using it in Examples
             output = gr.Label(
-                label="Analysis Result",
-                num_top_classes=2
+                label="📊 Analysis Result",
+                num_top_classes=2,
+                elem_id="output",
+                elem_classes="label-container"
             )
-    
-    # Prediction handling
-    submit_btn.click(
-        fn=predict,
-        inputs=[audio_input, model_dropdown],
-        outputs=[output]
+            
+            with gr.Accordion("ℹ️ How It Works", open=False):
+                gr.Markdown("""
+                    The SONICS classifier analyzes your audio to determine if it's an authentic song (Human created) or 
+                    generated by AI. Our models are trained on a diverse dataset of real and AI-generated songs from Suno and Udio.
+                    
+                    **Models available:**
+                    - **SpecTTTra-γ**: Optimized for speed
+                    - **SpecTTTra-β**: Balanced performance
+                    - **SpecTTTra-α**: Highest accuracy
+                    
+                    **Duration variants:**
+                    - **5s**: Analyzes a 5-second clip (faster)
+                    - **120s**: Analyzes up to 2 minutes (more accurate)
+                """)
+
+    # Add Examples section after output is defined
+    with gr.Accordion("🎬 Example Audio Files", open=True):
+        gr.Examples(
+            examples=[
+                ["example/real_song.mp3", "SpecTTTra-γ (5s)"],
+                ["example/fake_song.mp3", "SpecTTTra-γ (5s)"],
+            ],
+            inputs=[audio_input, model_dropdown],
+            outputs=[output],
+            fn=predict,
+            cache_examples=True,
+        )
+
+    # Footer
+    gr.HTML(
+        """
+        <div class="footer">
+            <p>SONICS: Synthetic Or Not - Identifying Counterfeit Songs | Created by SONICS Team</p>
+            <p>© 2025 - For research purposes only</p>
+        </div>
+        """
     )
 
+    # Prediction handling
+    submit_btn.click(fn=predict, inputs=[audio_input, model_dropdown], outputs=[output])
+
 if __name__ == "__main__":
     demo.launch()

example/fake_song.mp3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ba0ad7b7a7104a29ddf18c3ba3e04fb5045cdc1eb530f62fa611a08228eb30e
+size 4410477

example/real_song.mp3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c54d0a4d79601bdc739970ed8c22b6f5199527b79592146ebecf180e94f37529
+size 1922782