app.py

import gradio as gr
from transformers import pipeline
from youtube_transcript_api import YouTubeTranscriptApi
from youtube_transcript_api.formatters import TextFormatter
import re

# Define the models
models = {
    "Falconsai/text_summarization": "Falconsai/text_summarization",
    "suriya7/bart-finetuned-text-summarization": "suriya7/bart-finetuned-text-summarization"
}

# Default model
default_model = "Falconsai/text_summarization"

# Function to create a summarization pipeline
def create_summarization_pipeline(model_name):
    return pipeline("summarization", model=model_name)

# Function to extract video ID from URL
def extract_video_id(url):
    regex = r"(?:youtube\.com\/(?:[^\/\n\s]+\/\S+\/|(?:v|e(?:mbed)?)\/|\S*?[?&]v=)|youtu\.be\/)([a-zA-Z0-9_-]{11})"
    match = re.search(regex, url)
    if match:
        return match.group(1)
    return None

# Function to get YouTube transcript
def get_youtube_transcript(video_url):
    video_id = extract_video_id(video_url)
    if not video_id:
        return "Video ID could not be extracted. Please check the URL format."

    try:
        transcript = YouTubeTranscriptApi.get_transcript(video_id)
        formatter = TextFormatter()
        text_transcript = formatter.format_transcript(transcript)
        return text_transcript
    except Exception as e:
        error_message = str(e)
        if "Subtitles are disabled for this video" in error_message:
            return "Subtitles are disabled for this video. Transcript cannot be retrieved."
        return f"An error occurred while retrieving the transcript: {error_message}"

# Function to summarize YouTube video with selected model
def summarize_youtube_video(url, model_name):
    transcript = get_youtube_transcript(url)
    if "An error occurred" in transcript:
        return transcript

    # Truncate the transcript if necessary
    max_length = 1024  # Adjust according to the model's maximum sequence length
    if len(transcript) > max_length:
        transcript = transcript[:max_length]

    summarization_pipeline = create_summarization_pipeline(model_name)
    summary = summarization_pipeline(transcript, min_length=10, max_length=1000, do_sample=False)
    return summary[0]['summary_text']

# Define the Gradio interface
iface = gr.Interface(
    fn=summarize_youtube_video,
    inputs=[
        gr.Textbox(label="Enter YouTube Video URL", placeholder="e.g. https://www.youtube.com/watch?v=abcdef12345"),
        gr.Dropdown(choices=list(models.keys()), value=default_model, label="Select Summarization Model")
    ],
    outputs=gr.Textbox(label="Video Summary"),
    title="YouTube Video Summarizer",
    description="Enter the URL of a YouTube video and select a summarization model to get a summary of its transcript."
)

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