def parseFile(filename):
rArr=[]
try:
subs=SubRipFile.open(filename)
except UnicodeDecodeError:
subs=SubRipFile.open(filename,encoding='iso-8859-1')
for s in subs:
text=s.text
text=re.sub(fontpattern,'',text)
text=re.sub(npattern,'',text)
rArr.append(text)
return rArr
def parseString(string):
rArr=[]
subs=SubRipFile.from_string(string)
for s in subs:
text=s.text
text=re.sub(fontpattern,'',text)
text=re.sub(npattern,'',text)
rArr.append(text)
return rArr
def create_srt(split_df, cris_stt_df):
abs_path = os.path.dirname(split_df)
df1 = pd.read_csv(split_df)
df2 = pd.read_excel(cris_stt_df)
df1.rename(columns={'wav_filename': 'wav_name'}, inplace=True)
# This df3 contains all the info for srt creation
df3 = pd.merge(df1, df2, how='inner', on='wav_name')
print("Creating the srt:")
new_srt = SubRipFile()
for index, row in df3.iterrows():
text = str(row['transcripts'] if \
type(row['transcripts']) != float \
else "")
new_srt.append(
SubRipItem(index=index + 1,
start=SubRipTime(milliseconds=row['start']),
end=SubRipTime(milliseconds=row['end']),
text=text[:-1] if text.endswith(".") else text))
new_srt.save(os.path.join(abs_path, "stt_converted.srt"))
print("successfully written")
help="name of output file without extention (default: output)")
parser.add_argument('--eol', type=str, default="###",
help="End of line marker (default: ###)")
args=parser.parse_args()
subs = pysrt.open(args.file)
after_subs = []
indx = 1
iterator = iter(subs)
while True:
try:
sub = next(iterator)
except StopIteration:
break
if sub.text[-len(args.eol):] != args.eol:
after_subs.append(SubRipItem(indx, sub.start, sub.end, sub.text))
else:
line2_sub = next(iterator)
text = sub.text[:-len(args.eol)] + '\n' + line2_sub.text
combined_sub = SubRipItem(indx, sub.start, line2_sub.end, text=text)
after_subs.append(combined_sub)
indx += 1
after = SubRipFile(items=after_subs)
after.save(args.output + '.srt', encoding='utf-8')