Spaces:
Runtime error
Runtime error
ritwikraha
commited on
Commit
·
0c75fa8
1
Parent(s):
eed9b5d
add: first app
Browse files
app.py
ADDED
@@ -0,0 +1,67 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
# Import required libraries
|
2 |
+
import os
|
3 |
+
import re
|
4 |
+
import logging
|
5 |
+
import whisper
|
6 |
+
from pytube import YouTube
|
7 |
+
import gradio as gr
|
8 |
+
|
9 |
+
# Setup logging
|
10 |
+
logging.basicConfig(level=logging.INFO)
|
11 |
+
|
12 |
+
# Load the Whisper model
|
13 |
+
model = whisper.load_model("base")
|
14 |
+
|
15 |
+
def download_audio_from_youtube(url):
|
16 |
+
"""
|
17 |
+
Download the audio from a YouTube video and return the path to the audio file.
|
18 |
+
"""
|
19 |
+
yt = YouTube(url)
|
20 |
+
video = yt.streams.filter(only_audio=True).first()
|
21 |
+
out_file = video.download(output_path=".")
|
22 |
+
return out_file
|
23 |
+
|
24 |
+
def get_text(url):
|
25 |
+
"""
|
26 |
+
Transcribe the audio from a YouTube video and return the transcript.
|
27 |
+
"""
|
28 |
+
if not url:
|
29 |
+
return ''
|
30 |
+
|
31 |
+
out_file = download_audio_from_youtube(url)
|
32 |
+
file_stats = os.stat(out_file)
|
33 |
+
|
34 |
+
logging.info(f'Size of audio file in Bytes: {file_stats.st_size}')
|
35 |
+
|
36 |
+
if file_stats.st_size > 30000000:
|
37 |
+
logging.error('Videos for transcription on this space are limited to about 1.5 hours...')
|
38 |
+
return ''
|
39 |
+
|
40 |
+
base, ext = os.path.splitext(out_file)
|
41 |
+
new_file = base + '.mp3'
|
42 |
+
os.rename(out_file, new_file)
|
43 |
+
|
44 |
+
result = model.transcribe(new_file)
|
45 |
+
return result['text'].strip()
|
46 |
+
|
47 |
+
def create_gradio_interface():
|
48 |
+
"""
|
49 |
+
Create and launch a Gradio interface for transcribing YouTube videos.
|
50 |
+
"""
|
51 |
+
with gr.Blocks() as demo:
|
52 |
+
gr.Markdown("<h1><center>Free Fast YouTube URL Video-to-Text using <a href=https://openai.com/blog/whisper/ target=_blank>OpenAI's Whisper</a> Model</center></h1>")
|
53 |
+
gr.Markdown("<center>Enter the link of any YouTube video to generate a text transcript of the video.</center>")
|
54 |
+
gr.Markdown("<center><b>'Whisper is a neural net that approaches human level robustness and accuracy on English speech recognition.'</b></center>")
|
55 |
+
gr.Markdown("<center>Transcription takes 5-10 seconds per minute of the video. #patience<br />If you have time while waiting, check out my <a href=https://www.artificial-intelligence.blog target=_blank>AI blog</a> (opens in new tab).</center>")
|
56 |
+
|
57 |
+
input_text_url = gr.Textbox(placeholder='Youtube video URL', label='YouTube URL')
|
58 |
+
result_button_transcribe = gr.Button('Transcribe')
|
59 |
+
output_text_transcribe = gr.Textbox(placeholder='Transcript of the YouTube video.', label='Transcript')
|
60 |
+
|
61 |
+
result_button_transcribe.click(get_text, inputs=input_text_url, outputs=output_text_transcribe)
|
62 |
+
|
63 |
+
demo.queue(default_enabled=True).launch(debug=True)
|
64 |
+
|
65 |
+
# Launch the Gradio interface
|
66 |
+
if __name__ == "__main__":
|
67 |
+
create_gradio_interface()
|