def vtt_to_df(fn):
"""
Convert vtt to DataFrame
args:
fn - filepath to .vtt-file
returns:
DataFrame
"""
with open(fn) as f:
text = f.read()
vtt = WebVTTReader().read(text)
subtitles = []
for caption in vtt.get_captions('en-US'):
subtitles.append({
'time':
dt.datetime.strptime(caption.format_start(),
'%H:%M:%S.%f').strftime('%-Hh%mm%Ss'),
'start':
int((dt.datetime.strptime(caption.format_start(), '%H:%M:%S.%f') -
dt.datetime(1900, 1, 1)).total_seconds()),
'duration': (caption.end - caption.start) / 100000,
'text':
caption.get_text()
})
df = pd.DataFrame(subtitles)
return df
def getCaptions(url, progress_cb, so_far, task_weight):
ydl = youtube_dl.YoutubeDL({
'writesubtitles': True,
'allsubtitles': True,
'writeautomaticsub': True
})
with ydl:
res = ydl.extract_info(url, download=False)
if res['requested_subtitles'] and res['requested_subtitles']['en']:
print('Grabbing vtt file from ' +
res['requested_subtitles']['en']['url'])
response = requests.get(res['requested_subtitles']['en']['url'],
stream=True)
b = BytesIO()
for block in response.iter_content(1024):
b.write(block)
b.seek(0)
arr = WebVTTReader().read(b.read().decode('ascii'))
progress_cb(so_far + task_weight, so_far + task_weight)
return arr.get_captions('en-US')
else:
return []
print('Youtube Video does not have any english captions')
return None
new_captions = []
for s, sentence in enumerate(self.sentences):
for c, caption in enumerate(sentence.captions):
trans = match[s][c]
new_caption = deepcopy(caption.raw_caption)
new_caption.nodes = [CaptionNode.create_text(trans.strip())]
new_captions.append(new_caption)
# print(f'"{caption.raw_text}"', f'"{trans}"')
new_caption_set = CaptionSet({'en': new_captions})
return new_caption_set
input_file = Path("./sendung-vom-15112020-video-ut102~_type-webvtt.vtt")
read_srt = WebVTTReader().read(input_file.read_text('UTF-8'), lang='de')
sentence_manager = SentenceManager()
for raw_caption in read_srt.get_captions('de'):
caption = MyCaption(raw_caption)
sentence_manager.add_caption(caption)
# sentence_manager.finish()
# print(sentence_manager)
# sentence_manager.write_to_file(Path("./output.txt"))
match = sentence_manager.match_translation_from_file(
Path("./output_fixed.txt"), Path("./translated.txt"))
new_caption_set = sentence_manager.new_caption_set_from_match(match)
srt_output = SRTWriter().write(new_caption_set)
print(srt_output)
Path("./translated.srt").write_text(srt_output, 'UTF-8')
