Create main.py
Browse files
main.py
ADDED
@@ -0,0 +1,133 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import gradio as gr
|
2 |
+
from youtube_transcript_api import YouTubeTranscriptApi, TranscriptsDisabled, NoTranscriptFound
|
3 |
+
from youtube_transcript_api.formatters import TextFormatter, JSONFormatter, WebVTTFormatter, SRTFormatter
|
4 |
+
import json
|
5 |
+
|
6 |
+
def get_transcript(video_id, languages, format_type, translate_to, preserve_formatting):
|
7 |
+
try:
|
8 |
+
# Set default languages if none provided
|
9 |
+
if not languages:
|
10 |
+
languages = ['en']
|
11 |
+
else:
|
12 |
+
languages = languages.split(',')
|
13 |
+
|
14 |
+
# Get transcript
|
15 |
+
transcript = YouTubeTranscriptApi.get_transcript(
|
16 |
+
video_id,
|
17 |
+
languages=languages,
|
18 |
+
preserve_formatting=preserve_formatting
|
19 |
+
)
|
20 |
+
|
21 |
+
# Translate if requested
|
22 |
+
if translate_to:
|
23 |
+
transcript_list = YouTubeTranscriptApi.list_transcripts(video_id)
|
24 |
+
base_transcript = transcript_list.find_transcript(languages)
|
25 |
+
transcript = base_transcript.translate(translate_to).fetch()
|
26 |
+
|
27 |
+
# Format transcript based on selected format
|
28 |
+
formatter_map = {
|
29 |
+
"Text": TextFormatter(),
|
30 |
+
"JSON": JSONFormatter(),
|
31 |
+
"WebVTT": WebVTTFormatter(),
|
32 |
+
"SRT": SRTFormatter()
|
33 |
+
}
|
34 |
+
|
35 |
+
formatter = formatter_map[format_type]
|
36 |
+
formatted_transcript = formatter.format_transcript(transcript)
|
37 |
+
|
38 |
+
return formatted_transcript
|
39 |
+
|
40 |
+
except TranscriptsDisabled:
|
41 |
+
return "Error: Transcripts are disabled for this video"
|
42 |
+
except NoTranscriptFound:
|
43 |
+
return "Error: No transcript found for the specified languages"
|
44 |
+
except Exception as e:
|
45 |
+
return f"Unexpected error: {str(e)}"
|
46 |
+
|
47 |
+
def list_available_transcripts(video_id):
|
48 |
+
try:
|
49 |
+
transcript_list = YouTubeTranscriptApi.list_transcripts(video_id)
|
50 |
+
transcripts_info = []
|
51 |
+
|
52 |
+
for transcript in transcript_list:
|
53 |
+
info = {
|
54 |
+
"Language": transcript.language,
|
55 |
+
"Code": transcript.language_code,
|
56 |
+
"Is Generated": transcript.is_generated,
|
57 |
+
"Is Translatable": transcript.is_translatable,
|
58 |
+
"Translation Languages": transcript.translation_languages
|
59 |
+
}
|
60 |
+
transcripts_info.append(info)
|
61 |
+
|
62 |
+
return json.dumps(transcripts_info, indent=2)
|
63 |
+
except TranscriptsDisabled:
|
64 |
+
return "Error: Transcripts are disabled for this video"
|
65 |
+
except Exception as e:
|
66 |
+
return f"Error: {str(e)}"
|
67 |
+
|
68 |
+
# Create Gradio interface
|
69 |
+
with gr.Blocks(title="YouTube Transcript Fetcher") as demo:
|
70 |
+
gr.Markdown("# YouTube Transcript Fetcher")
|
71 |
+
gr.Markdown("Retrieve transcripts from YouTube videos with various formatting options")
|
72 |
+
|
73 |
+
with gr.Tab("Get Transcript"):
|
74 |
+
with gr.Row():
|
75 |
+
with gr.Column():
|
76 |
+
video_id_input = gr.Textbox(label="YouTube Video ID", placeholder="e.g., dQw4w9WgXcQ")
|
77 |
+
languages_input = gr.Textbox(
|
78 |
+
label="Languages (comma-separated)",
|
79 |
+
placeholder="e.g., en,de,es",
|
80 |
+
value="en"
|
81 |
+
)
|
82 |
+
format_dropdown = gr.Dropdown(
|
83 |
+
choices=["Text", "JSON", "WebVTT", "SRT"],
|
84 |
+
label="Output Format",
|
85 |
+
value="Text"
|
86 |
+
)
|
87 |
+
translate_dropdown = gr.Dropdown(
|
88 |
+
choices=["", "en", "de", "es", "fr", "it"],
|
89 |
+
label="Translate To (optional)",
|
90 |
+
value=""
|
91 |
+
)
|
92 |
+
preserve_formatting = gr.Checkbox(
|
93 |
+
label="Preserve Formatting",
|
94 |
+
value=False
|
95 |
+
)
|
96 |
+
submit_btn = gr.Button("Get Transcript")
|
97 |
+
|
98 |
+
with gr.Column():
|
99 |
+
output = gr.Textbox(label="Transcript", lines=20)
|
100 |
+
|
101 |
+
submit_btn.click(
|
102 |
+
fn=get_transcript,
|
103 |
+
inputs=[video_id_input, languages_input, format_dropdown, translate_dropdown, preserve_formatting],
|
104 |
+
outputs=output
|
105 |
+
)
|
106 |
+
|
107 |
+
with gr.Tab("List Available Transcripts"):
|
108 |
+
with gr.Row():
|
109 |
+
with gr.Column():
|
110 |
+
list_video_id = gr.Textbox(label="YouTube Video ID", placeholder="e.g., dQw4w9WgXcQ")
|
111 |
+
list_btn = gr.Button("List Transcripts")
|
112 |
+
|
113 |
+
with gr.Column():
|
114 |
+
list_output = gr.Textbox(label="Available Transcripts (JSON)", lines=20)
|
115 |
+
|
116 |
+
list_btn.click(
|
117 |
+
fn=list_available_transcripts,
|
118 |
+
inputs=list_video_id,
|
119 |
+
outputs=list_output
|
120 |
+
)
|
121 |
+
|
122 |
+
gr.Markdown("""
|
123 |
+
### Notes
|
124 |
+
- Enter a valid YouTube video ID (found in the URL)
|
125 |
+
- Specify languages as comma-separated codes (e.g., "en,de")
|
126 |
+
- Choose output format from available options
|
127 |
+
- Optional: Select a language to translate the transcript to
|
128 |
+
- Preserve formatting keeps HTML tags if present
|
129 |
+
""")
|
130 |
+
|
131 |
+
demo.launch(
|
132 |
+
share=True
|
133 |
+
)
|