def break_sentences(args, subs, alternative):
firstword = True
charcount = 0
idx = len(subs) + 1
content = ""
for w in alternative.words:
if firstword:
# first word in sentence, record start time
start = w.start_time.ToTimedelta()
charcount += len(w.word)
content += " " + w.word.strip()
if ("." in w.word or "!" in w.word or "?" in w.word
or charcount > args.max_chars
or ("," in w.word and not firstword)):
# break sentence at: . ! ? or line length exceeded
# also break if , and not first word
subs.append(
srt.Subtitle(index=idx,
start=start,
end=w.end_time.ToTimedelta(),
content=srt.make_legal_content(content)))
firstword = True
idx += 1
content = ""
charcount = 0
else:
firstword = False
return subs
def merge_srt(zh_file_path, en_file_path, new_file_path):
zh_file = open(zh_file_path, mode='r', encoding='utf-8')
zh_srt_s = zh_file.read()
zh_file.close()
en_file = open(en_file_path, mode='r', encoding='utf-8')
en_srt_s = en_file.read()
en_file.close()
zh_subs = list(srt.parse(zh_srt_s))
zh_length = len(zh_subs)
en_subs = list(srt.parse(en_srt_s))
en_length = len(en_subs)
if zh_length != en_length:
return False
subs = []
for i in range(zh_length):
temp_subtitle = zh_subs[i]
temp_subtitle.content = temp_subtitle.content + '\n' + en_subs[
i].content
temp_subtitle.content = srt.make_legal_content(temp_subtitle.content)
subs.append(temp_subtitle)
srt_s = srt.compose(subs)
# 将srt_s输出到new_file
new_file = open(new_file_path, 'w+', encoding='utf-8')
new_file.write(srt_s)
new_file.close()
return True
def say(self, string, time):
string = srt.make_legal_content(string.strip())
if string == '':
self.flush(time)
return
if string == self.string:
return
self.flush(time)
self.string = string
self.start = time
def generate_new_subtitle(self ):
subtitle_text = '\n'.join(self.caption)
subtitle_text = srt.make_legal_content(subtitle_text)
start = self.time_now
stop = start + self.time_delta
self.subtitles.append(
srt.Subtitle(len(self.subtitles)+1,
start,
stop,
subtitle_text)
)
self.time_now = stop
self.caption.clear()
def translate_srt(file_path, new_file_path, source, target):
# 从file中读取文本，以字符串形式存储到srt_s
file = open(file_path, mode='r', encoding='utf-8')
srt_s = file.read()
file.close()
# 将srt_s字符串转为包含Subtitle对象的列表subs
subs = list(srt.parse(srt_s))
length = len(subs)
# 将subs里Subtitle对象的content存储到ls，形成['字幕1','字幕2']的格式
ls = []
for i in range(length):
temp_content = re.sub(r'<\/?[\s\S]*?(?:".*")*>', '',
subs[i].content) # 去除所有<>标签
ls.append(temp_content)
ls_ = [] # 经过翻译的ls
# 写这一堆是因为单次请求的字符数不得大于2000
pre_index = 0
string = ''
for i in range(length):
if len(string + ls[i]) < 2000:
string = string + ls[i]
if i == length - 1:
temp_ls = ls[pre_index:]
ls_ = ls_ + translate(source, target, temp_ls)
elif len(ls[i]) >= 2000:
# print('该段字符数大于2000！')
break
else:
string = ''
temp_ls = ls[pre_index:i]
ls_ = ls_ + translate(source, target, temp_ls)
pre_index = i
# 更新subs里Subtitle对象的content
for i in range(length):
subs[i].content = srt.make_legal_content(ls_[i])
# subs转为字符串srt_s_
srt_s_ = srt.compose(subs)
# 将srt_s_输出到new_file
new_file = open(new_file_path, 'w+', encoding='utf-8')
new_file.write(srt_s_)
new_file.close()
def break_sentences(args, subs, alternative):
firstword = True
charcount = 0
idx = len(subs) + 1
content = ""
for w in alternative.words:
if firstword:
# first word in sentence, record start time
start_hhmmss = time.strftime('%H:%M:%S',
time.gmtime(w.start_time.seconds))
print("start_hhmmss: " + start_hhmmss)
start_ms = int(w.start_time.nanos / 1000000)
#print("start_ms: " + start_ms)
start = start_hhmmss + "," + str(start_ms)
print("start: " + start)
charcount += len(w.word)
content += " " + w.word.strip()
if ("." in w.word or "!" in w.word or "?" in w.word
or charcount > args.max_chars
or ("," in w.word and not firstword)):
# break sentence at: . ! ? or line length exceeded
# also break if , and not first word
end_hhmmss = time.strftime('%H:%M:%S',
time.gmtime(w.end_time.seconds))
print("end_hhmmss: " + end_hhmmss)
end_ms = int(w.end_time.nanos / 1000000)
end = end_hhmmss + "," + str(end_ms)
subs.append(
srt.Subtitle(index=idx,
start=srt.srt_timestamp_to_timedelta(start),
end=srt.srt_timestamp_to_timedelta(end),
content=srt.make_legal_content(content)))
firstword = True
idx += 1
content = ""
charcount = 0
else:
firstword = False
return subs
def merge_two_subtitles(index, sub_title_left, sub_title_right):
start = min(sub_title_left.start, sub_title_right.start)
end = max(sub_title_left.end, sub_title_right.end)
content = srt.make_legal_content('{}\n{}'.format(sub_title_left.content, sub_title_right.content))
return srt.Subtitle(index, start, end, content)
def get_lines(self):
lines = srt.parse(srt.make_legal_content(self.file.read()))
self.file.seek(0)
return lines
