Update app.py

app.py

@@ -30,36 +30,36 @@ if uploaded_file is not None:
     st.audio(uploaded_file)
 
     # Perform emotion classification
-    st.write("Classifying...")
-    try:
-        inputs = feature_extractor(audio_input, sampling_rate=sample_rate, return_tensors="pt")
-
-        # Make prediction
-        with torch.no_grad():
-            outputs = model(**inputs)
-
-        embeddings = outputs.pooler_output
-
-        # Apply a classification head on top of the embeddings
-        id2label={
-            0:"angry",
-            1:'calm',
-            2:'disgust',
-            3:'fearful',
-            4:'happy',
-            5:'neutral',
-            6:'sad',
-            7:'surprised'
-        }
-        classifier = torch.nn.Linear(embeddings.shape[-1], len(id2label))
-
-        # Pass embeddings through the classifier
-        logits = classifier(embeddings)
-
-        # Get predicted class
-        predicted_class_idx = logits.argmax(-1).item()
-        predicted_class = id2label[predicted_class_idx]
-
-        st.write(f"Predicted Emotion: {predicted_class}")
+    if st.button("Classifying"):
+        try:
+            inputs = feature_extractor(audio_input, sampling_rate=sample_rate, return_tensors="pt")
+    
+            # Make prediction
+            with torch.no_grad():
+                outputs = model(**inputs)
+    
+            embeddings = outputs.pooler_output
+    
+            # Apply a classification head on top of the embeddings
+            id2label={
+                0:"angry",
+                1:'calm',
+                2:'disgust',
+                3:'fearful',
+                4:'happy',
+                5:'neutral',
+                6:'sad',
+                7:'surprised'
+            }
+            classifier = torch.nn.Linear(embeddings.shape[-1], len(id2label))
+    
+            # Pass embeddings through the classifier
+            logits = classifier(embeddings)
+    
+            # Get predicted class
+            predicted_class_idx = logits.argmax(-1).item()
+            predicted_class = id2label[predicted_class_idx]
+    
+            st.write(f"Predicted Emotion: {predicted_class}")
     except Exception as e:
         st.write(f"Error during classification: {e}")