app.py

import streamlit as st
from transformers import BartForConditionalGeneration, BartTokenizer
from youtube_transcript_api import YouTubeTranscriptApi
from fastapi import FastAPI, Request
from fastapi.responses import JSONResponse
import uvicorn

# Initialize Streamlit app
st.title("YouTube Video Transcription Summarizer")
video_url = st.text_input("Enter YouTube Video URL:")

# Initialize FastAPI app
app = FastAPI()

# Load BART model and tokenizer
model_name = 'facebook/bart-large-cnn'
tokenizer = BartTokenizer.from_pretrained(model_name)
model = BartForConditionalGeneration.from_pretrained(model_name)

# Function to fetch transcript from YouTube URL
@st.cache
def get_transcript(url):
    try:
        video_id = url.split('=')[1]
        transcript_list = YouTubeTranscriptApi.get_transcript(video_id)
        transcript_text = ""
        for item in transcript_list:
            transcript_text += item['text'] + "\n"
        return transcript_text
    except Exception as e:
        return None

# Function to summarize transcript
@st.cache
def summarize_transcript(transcript):
    input_ids = tokenizer.encode("summarize: " + transcript, return_tensors="pt", max_length=1024, truncation=True)
    summary_ids = model.generate(input_ids, num_beams=4, min_length=30, max_length=200, early_stopping=True)
    summary = tokenizer.decode(summary_ids[0], skip_special_tokens=True)
    return summary

# API endpoint to summarize transcript
@app.post("/summarize_transcript")
async def summarize_transcript_api(request: Request):
    data = await request.json()
    video_url = data.get('video_url')
    transcript = get_transcript(video_url)
    if transcript:
        summary = summarize_transcript(transcript)
        return JSONResponse(content={"summary": summary})
    else:
        return JSONResponse(content={"error": "Error fetching transcript."}, status_code=400)

# Run Streamlit app
if __name__ == "__main__":
    uvicorn.run(app, host="0.0.0.0", port=8000)