app.py

from youtube_transcript_api import YouTubeTranscriptApi
from transformers import pipeline, AutoTokenizer
import torch
import re
import gradio as gr
import requests
import os

# -------------------
# CONFIGURATION
# -------------------
model_path = "facebook/bart-large-cnn"
max_tokens = 1024
YOUTUBE_API_KEY = "AIzaSyDCY9RM085oTxC8oms5z9TPzPKXKLFQAgc"  # Replace with your real API key

# -------------------
# Load model
# -------------------
tokenizer = AutoTokenizer.from_pretrained(model_path)
text_summary = pipeline(
    task="summarization",
    model=model_path,
    tokenizer=tokenizer,
    torch_dtype=torch.bfloat16 if torch.cuda.is_available() else torch.float32
)
from pydantic import BaseModel, PydanticUserError, ConfigDict
from pydantic import BaseModel, ConfigDict

class MyModel(BaseModel):
    request: 'starlette.requests.Request'
    model_config = ConfigDict(arbitrary_types_allowed=True)
from pydantic_core import core_schema
from starlette.requests import Request

def get_pydantic_core_schema(request_type, handler):
    return core_schema.any_schema()

Request.__get_pydantic_core_schema__ = get_pydantic_core_schema

# -------------------
# Utilities
# -------------------
def extract_video_id(url):
    regex = r"(?:v=|\/)([0-9A-Za-z_-]{11})"
    match = re.search(regex, url)
    if match:
        return match.group(1)
    else:
        raise ValueError("Invalid YouTube URL")

def fetch_video_metadata(video_id):
    try:
        url = f"https://www.googleapis.com/youtube/v3/videos?part=snippet&id={video_id}&key={YOUTUBE_API_KEY}"
        response = requests.get(url)
        data = response.json()
        if "items" in data and data["items"]:
            title = data["items"][0]["snippet"]["title"]
            description = data["items"][0]["snippet"]["description"]
            return title, description
        else:
            return "Title Not Found", "Description Not Found"
    except Exception as e:
        return "Error fetching title", str(e)

def get_transcript_text(video_id):
    try:
        transcript = YouTubeTranscriptApi.get_transcript(video_id)
        full_text = " ".join([entry['text'] for entry in transcript])
        return full_text
    except Exception as e:
        return None

def split_into_chunks(text, tokenizer, max_tokens):
    words = text.split()
    chunks = []
    current_chunk = []
    current_len = 0

    for word in words:
        token_len = len(tokenizer.tokenize(word))
        if current_len + token_len > max_tokens:
            chunks.append(" ".join(current_chunk))
            current_chunk = [word]
            current_len = token_len
        else:
            current_chunk.append(word)
            current_len += token_len

    if current_chunk:
        chunks.append(" ".join(current_chunk))
    return chunks

def summarize_text(full_text):
    if not full_text.strip():
        return "Transcript is empty or could not be retrieved."

    chunks = split_into_chunks(full_text, tokenizer, max_tokens)
    summaries = []
    for i, chunk in enumerate(chunks):
        try:
            print(f"Summarizing chunk {i+1}/{len(chunks)}...")
            summary = text_summary(chunk, max_length=180, min_length=10, do_sample=False)
            summaries.append(summary[0]['summary_text'])
        except Exception as e:
            summaries.append(f"[Error summarizing chunk: {str(e)}]")

    return "\n\n".join(summaries)

# -------------------
# Main Summary Function
# -------------------
def summarize_youtube_video(url):
    try:
        video_id = extract_video_id(url)
        title, description = fetch_video_metadata(video_id)
        transcript = get_transcript_text(video_id)
        if not transcript:
            return f"**Title**: {title}\n\n**Transcript not available.**"
        summary = summarize_text(transcript)
        return f"**Title**: {title}\n\n**Summary**:\n{summary}"
    except Exception as e:
        return f"Error: {str(e)}"

# -------------------
# Gradio UI
# -------------------
demo = gr.Interface(
    fn=summarize_youtube_video,
    inputs=[gr.Textbox(label='Enter YouTube URL')],
    outputs=[gr.Textbox(label='Video Title and Summary', lines=15)],
    title='YouTube Video Summarizer with Metadata',
    description='Paste a YouTube video URL to get a title and summarized content using transcript + YouTube API'
)

if __name__ == "__main__":
    demo.launch()