Update app.py

app.py

@@ -18,30 +18,48 @@ import sentencepiece  # For tokenization (required by SpeechT5Processor)
 ##########################################
 # Streamlit application title and input
 ##########################################
-st.title("Just Comment")  # Set the app title for user interface
-st.write("I'm listening to you, my friend")  # Add a brief app description
-text = st.text_area("Enter your comment", "")  # Text area for user to input their comment or feedback
+# Display a colorful, large title in a visually appealing font
+st.markdown(
+    "<h1 style='text-align: center; color: #FF5733; font-size: 50px;'>Just Comment</h1>",
+    unsafe_allow_html=True
+)  # Use HTML and CSS to set a custom title design
+
+# Display a smaller, gentle and 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
+)  # Use HTML to add a friendly and soft-styled subtitle
+
+# Add a well-designed text area for user input
+text = st.text_area(
+    "Enter your comment", 
+    placeholder="Type something here...", 
+    height=150, 
+    help="Write a comment you would like us to analyze and respond to!"  # Provide a helpful tooltip
+)
 
 ##########################################
 # Step 1: Sentiment Analysis Function
 ##########################################
 def analyze_dominant_emotion(user_review):
     """
-    Analyze the dominant emotion in the user's comment using our fine-tuned text classification model.
+    Analyze the dominant emotion in the user's comment using a fine-tuned text classification model.
     """
     emotion_classifier = pipeline(
         "text-classification", 
         model="Thea231/jhartmann_emotion_finetuning", 
         return_all_scores=True
-    )  # Load our fine-tuned text classification model
-    
-    emotion_results = emotion_classifier(user_review)[0]  # Get the emotion classification scores for the input text
-    dominant_emotion = max(emotion_results, key=lambda x: x['score'])  # Identify the emotion with the highest confidence
+    )  # Load the fine-tuned text classification model from Hugging Face
+
+    emotion_results = emotion_classifier(user_review)[0]  # Perform sentiment analysis on the input text
+    dominant_emotion = max(emotion_results, key=lambda x: x['score'])  # Identify the emotion with the highest confidence score
     return dominant_emotion  # Return the dominant emotion (label and score)
 
+
 ##########################################
 # Step 2: Response Generation Function
 ##########################################
+
 def response_gen(user_review):
     """
     Generate a concise and logical response based on the sentiment of the user's comment.
@@ -115,26 +133,26 @@ def response_gen(user_review):
         )
     }
     
-    # Select the appropriate prompt based on the user's emotion, or default to neutral
+    # Select the appropriate prompt based on the user's emotion or default to neutral
     prompt = emotion_prompts.get(
         emotion_label, 
         f"Neutral feedback: '{user_review}'\n\nWrite a professional and concise response (50-200 words max).\n\nResponse:"
     )
-    
-    # Load the tokenizer and language model for text generation
+
+    # Load the tokenizer and language model for response generation
     tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen1.5-0.5B")  # Load tokenizer for processing text inputs
-    model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen1.5-0.5B")  # Load language model for response generation
+    model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen1.5-0.5B")  # Load language model for text generation
 
     inputs = tokenizer(prompt, return_tensors="pt")  # Tokenize the input prompt
     outputs = model.generate(
         **inputs,
-        max_new_tokens=300,  # Set the upper limit of tokens generated to ensure the response isn't too lengthy
-        min_length=75,  # Set the minimum length of the generated response
-        no_repeat_ngram_size=2,  # Avoid repeating phrases
-        temperature=0.7  # Add slight randomness for natural-sounding responses
+        max_new_tokens=300,  # Set an upper limit on token generation to ensure concise output
+        min_length=75,  # Set a minimum length to ensure the response is complete
+        no_repeat_ngram_size=2,  # Avoid repetitive phrases
+        temperature=0.7  # Add randomness for more natural responses
     )
 
-    # Decode the generated response back into text
+    # Decode the generated response back into readable text
     response = tokenizer.decode(outputs[0], skip_special_tokens=True)
     print(f"Generated response: {response}")  # Print the response for debugging
     return response  # Return the generated response
@@ -161,10 +179,12 @@ def sound_gen(response):
     # Convert the spectrogram into an audio waveform using the vocoder
     with torch.no_grad():
         speech = vocoder(spectrogram)
-    
+
     # Save the audio as a .wav file
     sf.write("customer_service_response.wav", speech.numpy(), samplerate=16000)
-    st.audio("customer_service_response.wav")  # Embed an audio player in the web app
+
+    # Embed an auto-playing audio player in the web app
+    st.audio("customer_service_response.wav", start_time=0)  # Allow audio playback with autoplay feature
 
 ##########################################
 # Main Function
@@ -175,8 +195,11 @@ def main():
     """
     if text:  # Check if the user has entered a comment
         response = response_gen(text)  # Generate a logical and concise response
-        st.write(f"I wanna tell you that: {response}")  # Display the generated response in the Streamlit app
-        sound_gen(response)  # Convert the text response to speech and make it available for playback
+        st.markdown(
+            f"<p style='color:#2ECC71; font-size:20px;'>{response}</p>", 
+            unsafe_allow_html=True
+        )  # Display the generated response in a cute, styled font
+        sound_gen(response)  # Convert the response to speech and make it available for playback
 
 # Run the main function when the script is executed
 if __name__ == "__main__":