Update app.py

app.py

@@ -6,28 +6,38 @@ import numpy
 import scipy
 from gtts import gTTS
 from io import BytesIO
-from transformers import BartTokenizer
+from transformers import BartTokenizer, BartForConditionalGeneration
 
 def extract_text(pdf_file):
     pdfReader = PyPDF2.PdfReader(pdf_file)
     pageObj = pdfReader.pages[0]
     return pageObj.extract_text()
 
-
 def summarize_text(text):
     sentences = text.split(". ")
+    
+    # Find abstract section
     for i, sentence in enumerate(sentences):
         if "Abstract" in sentence:
             start = i + 1
             end = start + 3
             break
-    abstract = ". ".join(sentences[start:end+1]) 
+            
+    # Extract abstract 
+    abstract = ". ".join(sentences[start:end+1])
 
+    # Load BART model & tokenizer
     tokenizer = BartTokenizer.from_pretrained("facebook/bart-large-cnn")
-    summarizer = pipeline("summarization", model="facebook/bart-large-cnn", tokenizer=tokenizer) 
-    summary = summarizer(abstract, max_length=30, min_length=30,
-                         do_sample=False)
-    return summary[0]['summary_text']
+    model = BartForConditionalGeneration.from_pretrained("facebook/bart-large-cnn")
+
+    # Tokenize abstract 
+    inputs = tokenizer(abstract, return_tensors="pt", truncation=True)
+
+    # Generate summary
+    summary_ids = model.generate(inputs['input_ids'], num_beams=4, max_length=50, min_length=50, early_stopping=True)
+    summary = tokenizer.decode(summary_ids[0], skip_special_tokens=True)
+
+    return summary
 
 def text_to_audio(text):
     tts = gTTS(text, lang='en')