Update app.py

app.py

@@ -37,29 +37,23 @@ def sentiment_vader(sentence):
         overall_sentiment = "Neutral"
     return overall_sentiment
 
-# Function to transcribe audio and perform sentiment analysis
-def transcribe(audio):
-    time.sleep(3)  # Simulate processing delay
-    # In this case, just return a placeholder value
-    return "Transcription not available"
-
-# Function to get predictions for emotion and sentiment
+# Create a combined function that calls both models
 def get_predictions(audio_input):
     emotion_prediction = predict_emotion_from_audio(audio_input)
-    sentiment_prediction = sentiment_vader(transcribe(audio_input))
-    return emotion_prediction, sentiment_prediction
+    transcribe_prediction = transcribe(audio_input)
+    return [emotion_prediction, transcribe_prediction]
 
 # Create the Gradio interface
-interface = gr.Interface(
-    fn=get_predictions,
-    inputs=gr.Audio(label="Input Audio", type="file"),
-    outputs=[
-        gr.Label(label="Emotion Prediction"),
-        gr.Label(label="Sentiment Prediction")
-    ],
-    title="Emotional Machines Test",
-    description="Load an audio file to analyze speech emotion and sentiment."
-)
+with gr.Blocks() as interface:
+    gr.Markdown("Emotional Machines test: Load or Record an audio file to speech emotion analysis")
+    with gr.Tabs():
+        with gr.Tab("Acoustic and Semantic Predictions"):
+            with gr.Row():
+                input_audio = gr.Audio(label="Input Audio", type="filepath")
+                submit_button = gr.Button("Submit")
+            output_labels = [gr.Label(num_top_classes=8), gr.Label(num_top_classes=4)]
+
+    # Set the function to be called when the button is clicked
+    submit_button.click(get_predictions, inputs=input_audio, outputs=output_labels)
 
-# Launch the interface
 interface.launch()