Update app.py

app.py

@@ -13,13 +13,32 @@ from transformers import (
 from datasets import load_dataset  # For loading datasets (e.g., speaker embeddings)
 import torch  # For tensor operations
 import soundfile as sf  # For saving audio as .wav files
+import sentencepiece  # Required by SpeechT5Processor for tokenization
 
 ##########################################
 # Streamlit application title and input
 ##########################################
-st.title("Comment Reply for You")  # Application title
-st.write("Generate automatic replies for user comments")  # Application description
-text = st.text_area("Enter your comment", "")  # Text input for user to enter comments
+
+# Display a deep blue title in a large, visually appealing font
+st.markdown(
+    "<h1 style='text-align: center; color: #00008B; font-size: 50px;'>🚀 Just Comment</h1>",
+    unsafe_allow_html=True
+)  # Set deep blue title
+
+# Display a gentle, warm subtitle below the title
+st.markdown(
+    "<h3 style='text-align: center; color: #5D6D7E; font-style: italic;'>I'm listening to you, my friend～</h3>",
+    unsafe_allow_html=True
+)  # Set a friendly subtitle
+
+# Add a text area for user input with placeholder and tooltip
+text = st.text_area(
+    "Enter your comment",
+    placeholder="Type something here...",
+    height=100,
+    help="Write a comment you would like us to respond to!"  # Provide tooltip
+)  # Create text input field
+
 
 ##########################################
 # Step 1: Sentiment Analysis Function
@@ -32,11 +51,11 @@ def analyze_dominant_emotion(user_review):
         "text-classification", 
         model="Thea231/jhartmann_emotion_finetuning", 
         return_all_scores=True
-    )  # Load pre-trained emotion classification model
+    )  # Load our fine-tuned emotion classification model
     
     emotion_results = emotion_classifier(user_review)[0]  # Get emotion scores for the review
     dominant_emotion = max(emotion_results, key=lambda x: x['score'])  # Find the emotion with the highest confidence
-    return dominant_emotion
+    return dominant_emotion # Return the dominant emotion (as a dict with label and score)
 
 ##########################################
 # Step 2: Response Generation Function
@@ -45,11 +64,11 @@ def response_gen(user_review):
     """
     Generate a response based on the sentiment of the user's review.
     """
-    # Use Llama-based model to create a response based on a generated prompt
+    # Get dominant emotion for the input
     dominant_emotion = analyze_dominant_emotion(user_review)  # Get the dominant emotion
     emotion_label = dominant_emotion['label'].lower()  # Extract emotion label
     
-    # Define response templates for each emotion
+    # Define response templates for each emotion   
     emotion_prompts = {
         "anger": (
             "Customer complaint: '{review}'\n\n"
@@ -67,7 +86,53 @@ def response_gen(user_review):
             "- Invites them to explore loyalty programs\n\n"
             "Response:"
         ),
-        # Add other emotions as needed...
+        "disgust": (
+                "Customer quality concern: '{review}'\n\n"
+                "As a customer service representative, craft a response that:\n"
+                "- Immediately acknowledges the product issue.\n"
+                "- Explains measures taken in quality control.\n"
+                "- Provides clear return/replacement instructions.\n"
+                "- Offers a goodwill gesture (1-3 sentences).\n\n"
+                "Response:"
+        ),
+        "fear": (
+                "Customer safety concern: '{review}'\n\n"
+                "As a customer service representative, craft a reassuring response that:\n"
+                "- Directly addresses the safety worries.\n"
+                "- References relevant certifications or standards.\n"
+                "- Offers a dedicated support contact.\n"
+                "- Provides a satisfaction guarantee (1-3 sentences).\n\n"
+                "Response:"
+        ),
+        "neutral": (
+                "Customer feedback: '{review}'\n\n"
+                "As a customer service representative, craft a balanced response that:\n"
+                "- Provides additional relevant product information.\n"
+                "- Highlights key service features.\n"
+                "- Politely requests more detailed feedback.\n"
+                "- Maintains a professional tone (1-3 sentences).\n\n"
+                "Response:"
+        ),
+        "sadness": (
+                "Customer disappointment: '{review}'\n\n"
+                "As a customer service representative, craft an empathetic response that:\n"
+                "- Shows genuine understanding of the issue.\n"
+                "- Proposes a personalized recovery solution.\n"
+                "- Offers extended support options.\n"
+                "- Maintains a positive outlook (1-3 sentences).\n\n"
+                "Response:"
+        ),
+        "surprise": (
+                "Customer enthusiastic feedback: '{review}'\n\n"
+                "As a customer service representative, craft a response that:\n"
+                "- Matches the customer's positive energy.\n"
+                "- Highlights unexpected product benefits.\n"
+                "- Invites the customer to join community events or programs.\n"
+                "- Maintains the brand's voice (1-3 sentences).\n\n"
+                "Response:"
+        ),
+    
+        
     }
     
     # Format the prompt with the user's review