Update app.py

app.py

@@ -1,11 +1,22 @@
 import gradio as gr
 from PyPDF2 import PdfReader
 from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
+from gtts import gTTS
+from io import BytesIO
+import re
 
 model_name = "ArtifactAI/led_large_16384_arxiv_summarization"
 model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
 tokenizer = AutoTokenizer.from_pretrained(model_name)
 
+def extract_first_sentence(text):
+    # Use a simple regex to extract the first sentence
+    sentences = re.split(r'(?<!\w\.\w.)(?<![A-Z][a-z]\.)(?<=\.|\?)\s', text)
+    if sentences:
+        return sentences[0]
+    else:
+        return text
+
 def summarize_pdf_abstract(pdf_file):
     try:
         reader = PdfReader(pdf_file)
@@ -19,15 +30,26 @@ def summarize_pdf_abstract(pdf_file):
         outputs = model.generate(**inputs)
         summary = tokenizer.decode(outputs[0])
 
-        return {"summary": summary}
+        # Extract only the first sentence
+        summary_sentence = extract_first_sentence(summary)
+
+        # Generate audio
+        speech = gTTS(text=summary_sentence, lang="en")
+        speech_bytes = BytesIO()
+        speech.write_to_fp(speech_bytes)
+
+        return {"summary": summary_sentence, "audio": speech_bytes}
 
     except Exception as e:
         raise Exception(str(e))
 
+def play_audio(audio_bytes):
+    return gr.Audio(audio_bytes)
+
 interface = gr.Interface(
     fn=summarize_pdf_abstract,
     inputs=[gr.File(label="Upload PDF")],
-    outputs=[gr.Textbox(label="Summary")]
+    outputs=[gr.Textbox(label="Summary"), play_audio],
 )
 
 interface.launch(share=True)