Spaces:
Runtime error
Runtime error
Update app.py
Browse files
app.py
CHANGED
@@ -4,9 +4,20 @@ from youtube_transcript_api import YouTubeTranscriptApi
|
|
4 |
from youtube_transcript_api.formatters import TextFormatter
|
5 |
import re
|
6 |
|
7 |
-
#
|
8 |
-
|
|
|
|
|
|
|
9 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
10 |
def extract_video_id(url):
|
11 |
regex = r"(?:youtube\.com\/(?:[^\/\n\s]+\/\S+\/|(?:v|e(?:mbed)?)\/|\S*?[?&]v=)|youtu\.be\/)([a-zA-Z0-9_-]{11})"
|
12 |
match = re.search(regex, url)
|
@@ -14,6 +25,7 @@ def extract_video_id(url):
|
|
14 |
return match.group(1)
|
15 |
return None
|
16 |
|
|
|
17 |
def get_youtube_transcript(video_url):
|
18 |
video_id = extract_video_id(video_url)
|
19 |
if not video_id:
|
@@ -26,20 +38,25 @@ def get_youtube_transcript(video_url):
|
|
26 |
except Exception as e:
|
27 |
return f"An error occurred: {e}"
|
28 |
|
29 |
-
|
|
|
30 |
transcript = get_youtube_transcript(url)
|
31 |
if "An error occurred" in transcript:
|
32 |
return transcript
|
33 |
-
|
|
|
34 |
return summary[0]['summary_text']
|
35 |
|
36 |
# Define the Gradio interface
|
37 |
iface = gr.Interface(
|
38 |
fn=summarize_youtube_video,
|
39 |
-
inputs=
|
|
|
|
|
|
|
40 |
outputs=gr.Textbox(label="Video Summary"),
|
41 |
title="YouTube Video Summarizer",
|
42 |
-
description="Enter the URL of a YouTube video to get a summary of its transcript."
|
43 |
)
|
44 |
|
45 |
if __name__ == "__main__":
|
|
|
4 |
from youtube_transcript_api.formatters import TextFormatter
|
5 |
import re
|
6 |
|
7 |
+
# Define the models
|
8 |
+
models = {
|
9 |
+
"Falconsai/text_summarization": "Falconsai/text_summarization",
|
10 |
+
"suriya7/bart-finetuned-text-summarization": "suriya7/bart-finetuned-text-summarization"
|
11 |
+
}
|
12 |
|
13 |
+
# Default model
|
14 |
+
default_model = "Falconsai/text_summarization"
|
15 |
+
|
16 |
+
# Function to create a summarization pipeline
|
17 |
+
def create_summarization_pipeline(model_name):
|
18 |
+
return pipeline("summarization", model=model_name)
|
19 |
+
|
20 |
+
# Function to extract video ID from URL
|
21 |
def extract_video_id(url):
|
22 |
regex = r"(?:youtube\.com\/(?:[^\/\n\s]+\/\S+\/|(?:v|e(?:mbed)?)\/|\S*?[?&]v=)|youtu\.be\/)([a-zA-Z0-9_-]{11})"
|
23 |
match = re.search(regex, url)
|
|
|
25 |
return match.group(1)
|
26 |
return None
|
27 |
|
28 |
+
# Function to get YouTube transcript
|
29 |
def get_youtube_transcript(video_url):
|
30 |
video_id = extract_video_id(video_url)
|
31 |
if not video_id:
|
|
|
38 |
except Exception as e:
|
39 |
return f"An error occurred: {e}"
|
40 |
|
41 |
+
# Function to summarize YouTube video with selected model
|
42 |
+
def summarize_youtube_video(url, model_name):
|
43 |
transcript = get_youtube_transcript(url)
|
44 |
if "An error occurred" in transcript:
|
45 |
return transcript
|
46 |
+
summarization_pipeline = create_summarization_pipeline(model_name)
|
47 |
+
summary = summarization_pipeline(transcript, min_length=10, max_length=1000, do_sample=False)
|
48 |
return summary[0]['summary_text']
|
49 |
|
50 |
# Define the Gradio interface
|
51 |
iface = gr.Interface(
|
52 |
fn=summarize_youtube_video,
|
53 |
+
inputs=[
|
54 |
+
gr.Textbox(label="Enter YouTube Video URL", placeholder="e.g. https://www.youtube.com/watch?v=abcdef12345"),
|
55 |
+
gr.Dropdown(choices=list(models.keys()), value=default_model, label="Select Summarization Model")
|
56 |
+
],
|
57 |
outputs=gr.Textbox(label="Video Summary"),
|
58 |
title="YouTube Video Summarizer",
|
59 |
+
description="Enter the URL of a YouTube video and select a summarization model to get a summary of its transcript."
|
60 |
)
|
61 |
|
62 |
if __name__ == "__main__":
|