app.py

# Import necessary libraries
import gradio as gr
from youtube_transcript_api import YouTubeTranscriptApi
from transformers import pipeline

# Load a pre-trained summarization model from Hugging Face
# 'sshleifer/distilbart-cnn-12-6' is a good general-purpose summarization model
summarizer = pipeline("summarization", model="sshleifer/distilbart-cnn-12-6")

def get_youtube_id(url):
    """Extracts the YouTube video ID from a URL."""
    if "youtu.be/" in url:
        return url.split("youtu.be/")[1].split("?")[0]
    elif "v=" in url:
        return url.split("v=")[1].split("&")[0]
    return None

def summarize_youtube_transcript(youtube_url):
    """
    Fetches the transcript of a YouTube video and summarizes it.

    Args:
        youtube_url (str): The URL of the YouTube video.

    Returns:
        str: The summarized transcript or an error message.
    """
    video_id = get_youtube_id(youtube_url)

    if not video_id:
        return "Error: Could not extract YouTube video ID from the provided URL."

    try:
        # Get the transcript for the video ID
        transcript_list = YouTubeTranscriptApi.get_transcript(video_id)

        # Concatenate the transcript text
        transcript_text = " ".join([d['text'] for d in transcript_list])

        # Summarize the transcript
        # The summarizer pipeline handles splitting long texts if necessary,
        # but for very long videos, you might need more advanced chunking.
        # max_length and min_length control the summary length.
        summary = summarizer(transcript_text, max_length=200, min_length=50, do_sample=False)

        return summary[0]['summary_text']

    except Exception as e:
        return f"Error fetching or summarizing transcript: {e}"

# Create the Gradio interface
# The interface takes a text input (for the YouTube URL)
# and provides a text output (for the summarized transcript)
iface = gr.Interface(
    fn=summarize_youtube_transcript,
    inputs=gr.Textbox(label="Enter YouTube Video URL"),
    outputs=gr.Textbox(label="Summarized Transcript"),
    title="YouTube Transcript Summarizer",
    description="Enter a YouTube video URL to get a summary of its transcript."
)

# Launch the Gradio app
# share=True creates a temporary public link (useful for testing)
# Setting debug=True provides detailed logs
# iface.launch(share=True, debug=True)

# To deploy on Hugging Face Spaces, you just need this file (e.g., app.py)
# and a requirements.txt file. Hugging Face Spaces will automatically run
# the Gradio app if it finds an interface defined.
# Remove the iface.launch() call when deploying to Hugging Face Spaces.
# The last line should be the interface object itself.
iface.launch() # Use this line for local testing
# iface # Use this line for Hugging Face Spaces deployment