Spaces:
Sleeping
Sleeping
| import streamlit as st | |
| import pdfplumber | |
| from transformers import pipeline | |
| from gtts import gTTS | |
| import os | |
| # Function to extract text from PDF | |
| def extract_text_from_pdf(pdf_file): | |
| text = "" | |
| try: | |
| with pdfplumber.open(pdf_file) as pdf: | |
| for page in pdf.pages: | |
| text += page.extract_text() or "" | |
| except Exception as e: | |
| st.error(f"Error reading the PDF: {e}") | |
| return text | |
| # Function to generate discussion points with adjustable summary length | |
| def generate_discussion_points(text, user_prompt=None): | |
| summarizer = pipeline("summarization", model="facebook/bart-large-cnn") | |
| # Optionally include the user's prompt to guide the summary | |
| if user_prompt: | |
| text = user_prompt + " " + text | |
| # Adjust the max_length and min_length for a longer summary | |
| summary = summarizer(text, max_length=600, min_length=200, do_sample=False) | |
| return summary[0]['summary_text'] | |
| # Function to convert text to speech | |
| def text_to_speech(text, output_file="output.mp3"): | |
| tts = gTTS(text) | |
| tts.save(output_file) | |
| return output_file | |
| # Streamlit app starts here | |
| st.title("📄 PDF briefing Points Generator") | |
| st.write("Upload a PDF file to generate briefing points and listen to them. Optionally, provide a prompt to guide the summary.") | |
| # File uploader | |
| uploaded_file = st.file_uploader("Upload a PDF", type=["pdf"]) | |
| # Input for user-defined summary prompt | |
| user_prompt = st.text_area("Enter a prompt to guide the summary (optional)", height=100) | |
| if uploaded_file: | |
| # Extract text from uploaded PDF | |
| with st.spinner("Extracting text from the uploaded PDF..."): | |
| pdf_text = extract_text_from_pdf(uploaded_file) | |
| if pdf_text.strip(): | |
| st.write("### Extracted Text") | |
| st.text_area("Extracted Text", pdf_text, height=200) | |
| # Generate discussion points | |
| with st.spinner("Generating discussion points..."): | |
| try: | |
| discussion_points = generate_discussion_points(pdf_text, user_prompt) | |
| st.write("### Discussion Points") | |
| st.text_area("Discussion Points", discussion_points, height=150) | |
| except Exception as e: | |
| st.error(f"Error during summarization: {e}") | |
| # Convert discussion points to audio | |
| with st.spinner("Converting discussion points to audio..."): | |
| try: | |
| audio_file = text_to_speech(discussion_points) | |
| st.success("Audio file generated successfully!") | |
| except Exception as e: | |
| st.error(f"Error during text-to-speech conversion: {e}") | |
| audio_file = None | |
| if audio_file: | |
| # Audio playback | |
| st.write("### Listen to the Discussion Points") | |
| audio_bytes = open(audio_file, "rb").read() | |
| st.audio(audio_bytes, format="audio/mp3") | |
| # Option to download the audio file | |
| st.download_button( | |
| label="Download Audio", | |
| data=audio_bytes, | |
| file_name="discussion_points.mp3", | |
| mime="audio/mp3" | |
| ) | |
| else: | |
| st.error("No text was extracted from the uploaded PDF. Please try with another file.") | |
| else: | |
| st.info("Upload a PDF file to start.") | |