app.py

import gradio as gr
from gtts import gTTS
import pdfplumber
from sumy.parsers.plaintext import PlaintextParser
from sumy.nlp.tokenizers import Tokenizer
from sumy.summarizers.lsa import LsaSummarizer
import nltk
import os

# Download NLTK data for sumy
try:
    nltk.download('punkt')
    nltk.download('punkt_tab')
except Exception as e:
    print(f"Error downloading NLTK data: {str(e)}")

def extract_text_from_pdf(pdf_file):
    """
    Extract text from a PDF file using pdfplumber.
    
    Args:
        pdf_file: Uploaded PDF file.
    
    Returns:
        str: Extracted text or error message.
    """
    try:
        with pdfplumber.open(pdf_file) as pdf:
            text = ""
            for page in pdf.pages:
                page_text = page.extract_text()
                if page_text:
                    text += page_text + " "
        return text.strip() if text else "No text could be extracted from the PDF."
    except Exception as e:
        return f"Error extracting text: {str(e)}"

def summarize_text(text, sentences_count=12):
    """
    Summarize text to approximately four paragraphs using sumy LSA summarizer.
    
    Args:
        text (str): Text to summarize.
        sentences_count (int): Number of sentences in summary (approx. 3 sentences per paragraph).
    
    Returns:
        str: Summarized text or error message.
    """
    try:
        if len(text.split()) < 50:
            return "Text is too short to summarize."
        parser = PlaintextParser.from_string(text, Tokenizer("english"))
        summarizer = LsaSummarizer()
        summary = summarizer(parser.document, sentences_count)
        summary_text = ""
        for i, sentence in enumerate(summary):
            summary_text += str(sentence) + " "
            if (i + 1) % 3 == 0:
                summary_text += "\n\n"
        return summary_text.strip() if summary_text else "No summary generated."
    except Exception as e:
        return f"Error summarizing text: {str(e)}"

def pdf_to_speech(pdf_file, lang="en"):
    """
    Convert text from a PDF to summarized speech using gTTS.
    
    Args:
        pdf_file: Uploaded PDF file.
        lang (str): Language code (default is 'en' for English).
    
    Returns:
        tuple: (Path to audio file or None, summarized text or error message).
    """
    try:
        # Extract text from PDF
        text = extract_text_from_pdf(pdf_file)
        if "Error" in text:
            return None, text
        
        # Summarize text
        summarized_text = summarize_text(text, sentences_count=12)
        if "Error" in summarized_text or "too short" in summarized_text:
            return None, summarized_text
        
        # Create gTTS object
        tts = gTTS(text=summarized_text, lang=lang, slow=False)
        
        # Save the audio file
        output_file = "output.mp3"
        tts.save(output_file)
        
        return output_file, summarized_text
    
    except Exception as e:
        return None, f"An error occurred: {str(e)}"

# Define Gradio interface
demo = gr.Interface(
    fn=pdf_to_speech,
    inputs=[
        gr.File(label="Upload a PDF file", file_types=[".pdf"]),
        gr.Dropdown(choices=["en", "es", "fr"], label="Select Language", value="en")
    ],
    outputs=[
        gr.Audio(label="Generated Speech"),
        gr.Textbox(label="Summarized Text")
    ],
    title="PDF Summary to Speech",
    description="Upload an English PDF file, select a language, and generate speech from a summarized version (approx. 4 paragraphs). The summarized text is also displayed."
)

# Launch the app
if __name__ == "__main__":
    demo.launch()