Update app.py

app.py

@@ -125,11 +125,26 @@ Your audio is split into 5-second chunks, and each chunk is converted into a Mel
 A CNN predicts a label and confidence score for each chunk.
 
 The final prediction is determined by:
-1. **Majority vote** — the class predicted most frequently across chunks.  
-2. **Confidence tie-breaker** — if classes tie, the model selects the one with the **highest total confidence** across its chunks.  
+1. **Majority vote**, the class predicted most frequently across chunks.  
+2. **Confidence tie-breaker**, if classes tie, the model selects the one with the **highest total confidence** across its chunks.  
 3. **Final confidence** — the average confidence of all chunks belonging to the final class.
 
 The JSON output shows the final label, its confidence, and all per-chunk predictions.
+
+Classes you can classify them:
+. **Baby Cry**
+. **Chainsaw**
+. **Clock Tick**
+. **Cow**
+. **Dog**
+. **Fire Crackling**
+. **Frog**
+. **Helicopter**
+. **Person Sneeze**
+. **Pig**
+. **Rain**
+. **Rooster**
+. **Sea Waves**
 """
 
 # Gradio Interface