Update app.py
Browse files
app.py
CHANGED
@@ -5,6 +5,7 @@
|
|
5 |
|
6 |
# importing all necessary libraries
|
7 |
import httpcore
|
|
|
8 |
import pathlib
|
9 |
import sys, os
|
10 |
from gtts import gTTS
|
@@ -32,13 +33,13 @@ def download_video(url):
|
|
32 |
def validate_youtube(url):
|
33 |
#This creates a youtube objet
|
34 |
try:
|
35 |
-
yt = YouTube(url)
|
36 |
except Exception:
|
37 |
print("Hi there URL seems invalid")
|
38 |
return True
|
39 |
#This will return the length of the video in sec as an int
|
40 |
video_length = yt.length
|
41 |
-
if
|
42 |
print("Your video is larger than 10 minutes")
|
43 |
return True
|
44 |
else:
|
@@ -51,7 +52,7 @@ def validate_url(url):
|
|
51 |
print("Hi there URL seems invalid ")
|
52 |
return True
|
53 |
else:
|
54 |
-
return False
|
55 |
|
56 |
|
57 |
def cleanup():
|
@@ -62,16 +63,16 @@ def cleanup():
|
|
62 |
junks = []
|
63 |
for files in types:
|
64 |
junks.extend(glob.glob(files))
|
65 |
-
try:
|
66 |
# Deleting those files
|
67 |
for junk in junks:
|
68 |
print("Deleting",junk)
|
69 |
# Setting the path for the file to delete
|
70 |
file = pathlib.Path(junk)
|
71 |
# Calling the unlink method on the path
|
72 |
-
file.unlink()
|
73 |
except Exception:
|
74 |
-
print("I cannot delete the file because it is being used by another process")
|
75 |
|
76 |
def getSize(filename):
|
77 |
st = os.stat(filename)
|
@@ -82,33 +83,33 @@ def clean_transcript(transcript_list):
|
|
82 |
script = ""
|
83 |
for text in transcript_list:
|
84 |
t = text["text"]
|
85 |
-
if( (t != '[music]') and
|
86 |
-
(t != '[Music]') and
|
87 |
-
(t != '[музыка]') and
|
88 |
-
(t != '[Музыка]') and
|
89 |
-
(t != '[musik]') and
|
90 |
-
(t != '[Musik]') and
|
91 |
-
(t != '[musica]') and
|
92 |
-
(t != '[Musica]') and
|
93 |
-
(t != '[música]') and
|
94 |
-
(t != '[Música]') and
|
95 |
(t != '[音楽]') and \
|
96 |
-
(t != '[音乐]')
|
97 |
-
|
98 |
script += t + " "
|
99 |
return script
|
100 |
-
|
101 |
-
|
102 |
def get_transcript(url,desired_language):
|
103 |
id_you= url[url.index("=")+1:]
|
104 |
-
try:
|
105 |
# retrieve the available transcripts
|
106 |
transcript_list = YouTubeTranscriptApi.list_transcripts(id_you)
|
107 |
|
108 |
except Exception:
|
109 |
print('TranscriptsDisabled:')
|
110 |
is_translated = False
|
111 |
-
return " ", " ", is_translated
|
112 |
|
113 |
lista=[]
|
114 |
transcript_translation_languages=[]
|
@@ -130,12 +131,12 @@ def get_transcript(url,desired_language):
|
|
130 |
matrix=matrix.tolist()
|
131 |
is_manually=False
|
132 |
is_automatic=False
|
133 |
-
for lista in matrix:
|
134 |
#print(lista)
|
135 |
language_code=lista[0]
|
136 |
is_generated=lista[1]
|
137 |
is_translatable=lista[2]
|
138 |
-
if not is_generated and is_translatable :
|
139 |
print("Script found manually generated")
|
140 |
is_manually=True
|
141 |
language_code_man=language_code
|
@@ -143,12 +144,12 @@ def get_transcript(url,desired_language):
|
|
143 |
print("Script found automatic generated")
|
144 |
is_automatic=True
|
145 |
language_code_au=language_code
|
146 |
-
|
147 |
if is_manually:
|
148 |
# we try filter for manually created transcripts
|
149 |
print('We extract manually created transcripts')
|
150 |
-
transcript = transcript_list.find_manually_created_transcript([language_code])
|
151 |
-
|
152 |
elif is_automatic:
|
153 |
print('We extract generated transcript')
|
154 |
# or automatically generated ones, but not translated
|
@@ -168,13 +169,13 @@ def get_transcript(url,desired_language):
|
|
168 |
transcript_translated=transcript_translated.fetch()
|
169 |
translated=clean_transcript(transcript_translated)
|
170 |
is_translated = True
|
171 |
-
script_translated = ""
|
172 |
if is_translated :
|
173 |
script_translated = translated
|
174 |
|
175 |
transcript=transcript.fetch()
|
176 |
script = clean_transcript(transcript)
|
177 |
-
|
178 |
return script, script_translated, is_translated
|
179 |
|
180 |
# Set environment variables
|
@@ -225,7 +226,7 @@ def video_to_translate(url,initial_language,final_language):
|
|
225 |
elif final_language == "Vietnamese":
|
226 |
lang='vi'
|
227 |
elif final_language == "Japanese":
|
228 |
-
lang='ja'
|
229 |
# Initial directory
|
230 |
home_dir= os.getenv('home_dir')
|
231 |
print('Initial directory:',home_dir)
|
@@ -271,15 +272,15 @@ def video_to_translate(url,initial_language,final_language):
|
|
271 |
text=text+text_chunk+" "
|
272 |
text=str(text)
|
273 |
print(type(text))
|
274 |
-
|
275 |
else:
|
276 |
try:
|
277 |
text = r.recognize_google(audio_data, language = lang_in)
|
278 |
except Exception:
|
279 |
-
|
280 |
-
|
281 |
-
|
282 |
-
|
283 |
#print(text)
|
284 |
print("Destination language ",lang)
|
285 |
|
@@ -293,20 +294,20 @@ def video_to_translate(url,initial_language,final_language):
|
|
293 |
print("This text cannot be translated")
|
294 |
cleanup()
|
295 |
return "./demo/tryagain.mp4"
|
296 |
-
|
297 |
#translation.text
|
298 |
trans=translation.text
|
299 |
|
300 |
-
myobj = gTTS(text=trans, lang=lang, slow=False)
|
301 |
-
myobj.save("audio.wav")
|
302 |
# loading audio file
|
303 |
audioclip = AudioFileClip("audio.wav")
|
304 |
-
|
305 |
# adding audio to the video clip
|
306 |
new_audioclip = CompositeAudioClip([audioclip])
|
307 |
videoclip.audio = new_audioclip
|
308 |
new_video="video_translated_"+lang+".mp4"
|
309 |
-
|
310 |
# Return back to main directory
|
311 |
os.chdir(home_dir)
|
312 |
print('Final directory',os.getcwd())
|
@@ -326,7 +327,7 @@ gr.Interface(
|
|
326 |
fn=video_to_translate,
|
327 |
inputs=[url, initial_language, final_language],
|
328 |
outputs='video',
|
329 |
-
#verbose=True, #
|
330 |
title='Video Youtube Translator',
|
331 |
description='A simple application that translates Youtube small videos from English, Italian, Japanese, Russian, Spanish, and German to Italian, Spanish, Russian, English, and Japanese. Wait one minute to process.',
|
332 |
article='''<div>
|
@@ -335,11 +336,6 @@ gr.Interface(
|
|
335 |
</p>
|
336 |
</div>''',
|
337 |
examples=[
|
338 |
-
["https://www.youtube.com/watch?v=uLVRZE8OAI4", "English", "Spanish"]
|
339 |
-
["https://www.youtube.com/watch?v=fkGCLIQx1MI", "English", "Russian"],
|
340 |
-
["https://www.youtube.com/watch?v=6Q6hFtitthQ", "Italian", "English"],
|
341 |
-
["https://www.youtube.com/watch?v=s5XvjAC7ai8", "Russian", "English"],
|
342 |
-
["https://www.youtube.com/watch?v=qzzweIQoIOU", "Japanese", "English"],
|
343 |
-
["https://www.youtube.com/watch?v=nOGZvu6tJFE", "German", "Spanish"]
|
344 |
]
|
345 |
).launch()
|
|
|
5 |
|
6 |
# importing all necessary libraries
|
7 |
import httpcore
|
8 |
+
#setattr(httpcore, 'SyncHTTPTransport', 'AsyncHTTPProxy') # This line is problematic and should be removed or corrected based on the intended use.
|
9 |
import pathlib
|
10 |
import sys, os
|
11 |
from gtts import gTTS
|
|
|
33 |
def validate_youtube(url):
|
34 |
#This creates a youtube objet
|
35 |
try:
|
36 |
+
yt = YouTube(url)
|
37 |
except Exception:
|
38 |
print("Hi there URL seems invalid")
|
39 |
return True
|
40 |
#This will return the length of the video in sec as an int
|
41 |
video_length = yt.length
|
42 |
+
if video_length > 600:
|
43 |
print("Your video is larger than 10 minutes")
|
44 |
return True
|
45 |
else:
|
|
|
52 |
print("Hi there URL seems invalid ")
|
53 |
return True
|
54 |
else:
|
55 |
+
return False
|
56 |
|
57 |
|
58 |
def cleanup():
|
|
|
63 |
junks = []
|
64 |
for files in types:
|
65 |
junks.extend(glob.glob(files))
|
66 |
+
try:
|
67 |
# Deleting those files
|
68 |
for junk in junks:
|
69 |
print("Deleting",junk)
|
70 |
# Setting the path for the file to delete
|
71 |
file = pathlib.Path(junk)
|
72 |
# Calling the unlink method on the path
|
73 |
+
file.unlink()
|
74 |
except Exception:
|
75 |
+
print("I cannot delete the file because it is being used by another process")
|
76 |
|
77 |
def getSize(filename):
|
78 |
st = os.stat(filename)
|
|
|
83 |
script = ""
|
84 |
for text in transcript_list:
|
85 |
t = text["text"]
|
86 |
+
if( (t != '[music]') and \
|
87 |
+
(t != '[Music]') and \
|
88 |
+
(t != '[музыка]') and \
|
89 |
+
(t != '[Музыка]') and \
|
90 |
+
(t != '[musik]') and \
|
91 |
+
(t != '[Musik]') and \
|
92 |
+
(t != '[musica]') and \
|
93 |
+
(t != '[Musica]') and \
|
94 |
+
(t != '[música]') and \
|
95 |
+
(t != '[Música]') and \
|
96 |
(t != '[音楽]') and \
|
97 |
+
(t != '[音乐]')
|
98 |
+
) :
|
99 |
script += t + " "
|
100 |
return script
|
101 |
+
|
102 |
+
|
103 |
def get_transcript(url,desired_language):
|
104 |
id_you= url[url.index("=")+1:]
|
105 |
+
try:
|
106 |
# retrieve the available transcripts
|
107 |
transcript_list = YouTubeTranscriptApi.list_transcripts(id_you)
|
108 |
|
109 |
except Exception:
|
110 |
print('TranscriptsDisabled:')
|
111 |
is_translated = False
|
112 |
+
return " ", " ", is_translated
|
113 |
|
114 |
lista=[]
|
115 |
transcript_translation_languages=[]
|
|
|
131 |
matrix=matrix.tolist()
|
132 |
is_manually=False
|
133 |
is_automatic=False
|
134 |
+
for lista in matrix:
|
135 |
#print(lista)
|
136 |
language_code=lista[0]
|
137 |
is_generated=lista[1]
|
138 |
is_translatable=lista[2]
|
139 |
+
if not is_generated and is_translatable :
|
140 |
print("Script found manually generated")
|
141 |
is_manually=True
|
142 |
language_code_man=language_code
|
|
|
144 |
print("Script found automatic generated")
|
145 |
is_automatic=True
|
146 |
language_code_au=language_code
|
147 |
+
|
148 |
if is_manually:
|
149 |
# we try filter for manually created transcripts
|
150 |
print('We extract manually created transcripts')
|
151 |
+
transcript = transcript_list.find_manually_created_transcript([language_code])
|
152 |
+
|
153 |
elif is_automatic:
|
154 |
print('We extract generated transcript')
|
155 |
# or automatically generated ones, but not translated
|
|
|
169 |
transcript_translated=transcript_translated.fetch()
|
170 |
translated=clean_transcript(transcript_translated)
|
171 |
is_translated = True
|
172 |
+
script_translated = ""
|
173 |
if is_translated :
|
174 |
script_translated = translated
|
175 |
|
176 |
transcript=transcript.fetch()
|
177 |
script = clean_transcript(transcript)
|
178 |
+
|
179 |
return script, script_translated, is_translated
|
180 |
|
181 |
# Set environment variables
|
|
|
226 |
elif final_language == "Vietnamese":
|
227 |
lang='vi'
|
228 |
elif final_language == "Japanese":
|
229 |
+
lang='ja'
|
230 |
# Initial directory
|
231 |
home_dir= os.getenv('home_dir')
|
232 |
print('Initial directory:',home_dir)
|
|
|
272 |
text=text+text_chunk+" "
|
273 |
text=str(text)
|
274 |
print(type(text))
|
275 |
+
|
276 |
else:
|
277 |
try:
|
278 |
text = r.recognize_google(audio_data, language = lang_in)
|
279 |
except Exception:
|
280 |
+
print("This video cannot be recognized")
|
281 |
+
cleanup()
|
282 |
+
return "./demo/tryagain.mp4"
|
283 |
+
|
284 |
#print(text)
|
285 |
print("Destination language ",lang)
|
286 |
|
|
|
294 |
print("This text cannot be translated")
|
295 |
cleanup()
|
296 |
return "./demo/tryagain.mp4"
|
297 |
+
|
298 |
#translation.text
|
299 |
trans=translation.text
|
300 |
|
301 |
+
myobj = gTTS(text=trans, lang=lang, slow=False)
|
302 |
+
myobj.save("audio.wav")
|
303 |
# loading audio file
|
304 |
audioclip = AudioFileClip("audio.wav")
|
305 |
+
|
306 |
# adding audio to the video clip
|
307 |
new_audioclip = CompositeAudioClip([audioclip])
|
308 |
videoclip.audio = new_audioclip
|
309 |
new_video="video_translated_"+lang+".mp4"
|
310 |
+
|
311 |
# Return back to main directory
|
312 |
os.chdir(home_dir)
|
313 |
print('Final directory',os.getcwd())
|
|
|
327 |
fn=video_to_translate,
|
328 |
inputs=[url, initial_language, final_language],
|
329 |
outputs='video',
|
330 |
+
#verbose=True, # Removed verbose as it's no longer a valid parameter
|
331 |
title='Video Youtube Translator',
|
332 |
description='A simple application that translates Youtube small videos from English, Italian, Japanese, Russian, Spanish, and German to Italian, Spanish, Russian, English, and Japanese. Wait one minute to process.',
|
333 |
article='''<div>
|
|
|
336 |
</p>
|
337 |
</div>''',
|
338 |
examples=[
|
339 |
+
["https://www.youtube.com/watch?v=uLVRZE8OAI4", "English", "Spanish"]
|
|
|
|
|
|
|
|
|
|
|
340 |
]
|
341 |
).launch()
|