app.py

import gradio as gr
from PyPDF2 import PdfReader
from transformers import pipeline
from gtts import gTTS
import tempfile
import os

def extract_abstract_from_pdf(file):
    reader = PdfReader(file.name)
    first_page = reader.pages[0]
    text = first_page.extract_text()
    return text

def summarize_text(text, model_name='sshleifer/distilbart-cnn-12-6'):
    summarizer = pipeline("summarization", model=model_name)
    summary = summarizer(text, max_length=75, min_length=20, do_sample=False)
    first_sentence = summary[0]['summary_text'].split('.')[0] + '.'
    return first_sentence

def process_pdf_to_speech(file):
    abstract = extract_abstract_from_pdf(file)
    summary = summarize_text(abstract)
    tts = gTTS(text=summary, lang='en')
    with tempfile.NamedTemporaryFile(delete=False, suffix='.mp3') as fp:
        tts.save(fp.name)
        return fp.name

iface = gr.Interface(
    fn=process_pdf_to_speech,
    inputs=gr.File(label="Upload PDF"),
    outputs=gr.outputs.Audio(type="file", label="Play Summary"),
    title="PDF Abstract to Speech",
    description="This app summarizes the abstract from a PDF and converts it to speech. Please upload a PDF with an abstract."
)

iface.launch()