Update app.py

app.py

@@ -4,31 +4,53 @@ from langdetect.lang_detect_exception import LangDetectException
 import streamlit as st
 from transformers import pipeline
 
-# Initialize the pipelines
-summarizer = pipeline("summarization", model="Falconsai/text_summarization")
-classifier = pipeline("text-classification", model='Rainess/Finturned-Music-Sentiment', return_all_scores=True)
-
-@app.route('/process_lyrics', methods=['POST'])
-def process_lyrics():
-    data = request.json
-    lyrics = data.get('lyrics')
-
-    if not lyrics:
-        return jsonify({'error': 'No lyrics provided'}), 400
 
-    # Summarize the lyrics
-    summary = summarizer(lyrics, max_length=50, min_length=25, do_sample=False)[0]['summary_text']
-
-    # Classify the summarized text
-    classification_result = classifier(summary)[0]
+# Load the text summarization model pipeline
+summarizer = pipeline("summarization", model="Falconsai/text_summarization")
 
-    # Determine the sentiment
-    sentiment = "Negative" if classification_result['label'] == 'LABEL_0' else "Positive"
+# Load the text classification model pipeline
+classifier = pipeline("text-classification", model='Rainess/Finturned-Music-Sentiment', return_all_scores=True)
 
-    return jsonify({
-        'summary': summary,
-        'sentiment': sentiment
-    })
+def summary_song(text):
+    input_length = len(text)
+    if input_length < 200:
+        return text  # Return original text if length is less than 200
+    else:
+        max_length = 200
+        min_length = 50
+        summary = summarizer(text, max_length=max_length, min_length=min_length, truncation=True)
+        summarized_text = summary[0]['summary_text']
+        return summarized_text
+
+# Streamlit application title
+st.title("Text Summarization and Sentiment Classification for Music Lyrics")
+st.write("Classify the sentiment of summarized lyrics as Positive or Negative")
+
+# Text input for user to enter the lyrics
+text = st.text_area("Enter the lyrics to summarize and classify", "")
+
+# Perform text summarization and classification when the user clicks the "Summarize and Classify" button
+if st.button("Summarize and Classify"):
+    if text.strip():
+        # Perform text summarization on the input lyrics
+        summarized_text = summary_song(text)
+        
+        # Perform text classification on the summarized text
+        classification_results = classifier(summarized_text)
+        
+        # Get the highest scoring sentiment
+        max_score = float('-inf')
+        max_label = ''
+        
+        
+        # Map label to sentiment
+        sentiment = "Positive" if max_label == "1" else "Negative"
+        
+        # Display the summarized text and classification result
+        st.write("Original Lyrics:", text)
+        st.write("Summarized Text:", summarized_text)
+        st.write("Sentiment:", sentiment)
+        st.write("Score:", max_score)
+    else:
+        st.write("Please enter some lyrics to summarize and classify.")
 
-if __name__ == '__main__':
-    app.run(host='0.0.0.0', port=5000)