adarshjha01
commited on
Create app.py
Browse files
app.py
ADDED
@@ -0,0 +1,65 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
|
2 |
+
from youtube_transcript_api import YouTubeTranscriptApi
|
3 |
+
from youtube_transcript_api.formatters import TextFormatter
|
4 |
+
import re
|
5 |
+
import torch
|
6 |
+
import gradio as gr
|
7 |
+
from transformers import pipeline
|
8 |
+
|
9 |
+
# Initialize summarization pipeline
|
10 |
+
text_summary = pipeline("summarization", model="sshleifer/distilbart-cnn-12-6",
|
11 |
+
torch_dtype=torch.bfloat16)
|
12 |
+
|
13 |
+
def summarize_text(input_text):
|
14 |
+
"""Summarizes the given text."""
|
15 |
+
if not input_text.strip():
|
16 |
+
return "No transcript available to summarize."
|
17 |
+
output = text_summary(input_text, max_length=200, min_length=30, do_sample=False)
|
18 |
+
return output[0]['summary_text']
|
19 |
+
|
20 |
+
def get_video_id(url):
|
21 |
+
"""Extracts video ID from a YouTube URL."""
|
22 |
+
match = re.search(r"(?:v=|\/)([0-9A-Za-z_-]{11})", url)
|
23 |
+
if not match:
|
24 |
+
raise ValueError("Invalid YouTube URL")
|
25 |
+
return match.group(1)
|
26 |
+
|
27 |
+
def fetch_transcript(video_id):
|
28 |
+
"""
|
29 |
+
Fetches the transcript of the YouTube video in English or Hindi.
|
30 |
+
Returns the transcript text and the language.
|
31 |
+
"""
|
32 |
+
languages_to_try = [["en"], ["hi"]]
|
33 |
+
for language_codes in languages_to_try:
|
34 |
+
try:
|
35 |
+
transcript = YouTubeTranscriptApi.get_transcript(video_id, languages=language_codes)
|
36 |
+
formatter = TextFormatter()
|
37 |
+
return formatter.format_transcript(transcript), language_codes[0]
|
38 |
+
except Exception as e:
|
39 |
+
continue # Try the next language in the list
|
40 |
+
raise ValueError("Transcript not available in English or Hindi.")
|
41 |
+
|
42 |
+
def process_youtube_url(url):
|
43 |
+
"""Fetches the transcript from the YouTube URL and summarizes it."""
|
44 |
+
try:
|
45 |
+
video_id = get_video_id(url)
|
46 |
+
transcript, language = fetch_transcript(video_id)
|
47 |
+
language_name = "English" if language == "en" else "Hindi"
|
48 |
+
summary = summarize_text(transcript)
|
49 |
+
return f"Transcript Language: {language_name}\n\nSummary:\n{summary}"
|
50 |
+
except ValueError as e:
|
51 |
+
return str(e)
|
52 |
+
except Exception as e:
|
53 |
+
return f"An unexpected error occurred: {e}"
|
54 |
+
|
55 |
+
# Gradio Interface
|
56 |
+
demo = gr.Interface(
|
57 |
+
fn=process_youtube_url,
|
58 |
+
inputs=[gr.Textbox(label="Enter YouTube URL", placeholder="https://www.youtube.com/watch?v=example")],
|
59 |
+
outputs=[gr.Textbox(label="Summary")],
|
60 |
+
title="YouTube Video Transcript Summarizer",
|
61 |
+
description="Enter a YouTube video URL to generate it's summary. The system will first attempt to fetch the transcript in English, then Hindi if English is unavailable."
|
62 |
+
)
|
63 |
+
|
64 |
+
if __name__ == "__main__":
|
65 |
+
demo.launch()
|