mirror of
https://github.com/RVC-Boss/GPT-SoVITS.git
synced 2025-10-09 08:20:00 +08:00
269 lines
8.5 KiB
Python
269 lines
8.5 KiB
Python
|
||
|
||
|
||
|
||
import re
|
||
from typing import Callable
|
||
from tools.i18n.i18n import I18nAuto
|
||
|
||
i18n = I18nAuto()
|
||
|
||
METHODS = dict()
|
||
|
||
def get_method(name:str)->Callable:
|
||
method = METHODS.get(name, None)
|
||
if method is None:
|
||
raise ValueError(f"Method {name} not found")
|
||
return method
|
||
|
||
def register_method(name):
|
||
def decorator(func):
|
||
METHODS[name] = func
|
||
return func
|
||
return decorator
|
||
|
||
splits = {"，", "。", "？", "！", ",", ".", "?", "!", "~", ":", "：", "—", "…", }
|
||
|
||
def split_big_text(text, max_len=510):
|
||
# 定义全角和半角标点符号
|
||
punctuation = "".join(splits)
|
||
|
||
# 切割文本
|
||
segments = re.split('([' + punctuation + '])', text)
|
||
|
||
# 初始化结果列表和当前片段
|
||
result = []
|
||
current_segment = ''
|
||
|
||
for segment in segments:
|
||
# 如果当前片段加上新的片段长度超过max_len，就将当前片段加入结果列表，并重置当前片段
|
||
if len(current_segment + segment) > max_len:
|
||
result.append(current_segment)
|
||
current_segment = segment
|
||
else:
|
||
current_segment += segment
|
||
|
||
# 将最后一个片段加入结果列表
|
||
if current_segment:
|
||
result.append(current_segment)
|
||
|
||
return result
|
||
|
||
|
||
|
||
def split(todo_text):
|
||
todo_text = todo_text.replace("……", "。").replace("——", "，")
|
||
if todo_text[-1] not in splits:
|
||
todo_text += "。"
|
||
i_split_head = i_split_tail = 0
|
||
len_text = len(todo_text)
|
||
todo_texts = []
|
||
while 1:
|
||
if i_split_head >= len_text:
|
||
break # 结尾一定有标点，所以直接跳出即可，最后一段在上次已加入
|
||
if todo_text[i_split_head] in splits:
|
||
i_split_head += 1
|
||
todo_texts.append(todo_text[i_split_tail:i_split_head])
|
||
i_split_tail = i_split_head
|
||
else:
|
||
i_split_head += 1
|
||
return todo_texts
|
||
|
||
|
||
# 不切
|
||
@register_method("cut0")
|
||
def cut0(inp):
|
||
return inp
|
||
|
||
|
||
# 凑四句一切
|
||
@register_method("cut1")
|
||
def cut1(inp):
|
||
inp = inp.strip("\n")
|
||
inps = split(inp)
|
||
split_idx = list(range(0, len(inps), 4))
|
||
# split_idx[-1] = None
|
||
split_idx.append(None)
|
||
if len(split_idx) > 1:
|
||
opts = []
|
||
for idx in range(len(split_idx) - 1):
|
||
opts.append("".join(inps[split_idx[idx]: split_idx[idx + 1]]))
|
||
else:
|
||
opts = [inp]
|
||
return "\n".join(opts)
|
||
|
||
# 凑50字一切
|
||
@register_method("cut2")
|
||
def cut2(inp):
|
||
inp = inp.strip("\n")
|
||
inps = split(inp)
|
||
if len(inps) < 2:
|
||
return inp
|
||
opts = []
|
||
summ = 0
|
||
tmp_str = ""
|
||
for i in range(len(inps)):
|
||
summ += len(inps[i])
|
||
tmp_str += inps[i]
|
||
if summ > 50:
|
||
summ = 0
|
||
opts.append(tmp_str)
|
||
tmp_str = ""
|
||
if tmp_str != "":
|
||
opts.append(tmp_str)
|
||
# print(opts)
|
||
if len(opts) > 1 and len(opts[-1]) < 50: ##如果最后一个太短了，和前一个合一起
|
||
opts[-2] = opts[-2] + opts[-1]
|
||
opts = opts[:-1]
|
||
return "\n".join(opts)
|
||
|
||
# 按中文句号。切
|
||
@register_method("cut3")
|
||
def cut3(inp):
|
||
inp = inp.strip("\n")
|
||
return "\n".join(["%s" % item for item in inp.strip("。").split("。")])
|
||
|
||
#按英文句号.切
|
||
@register_method("cut4")
|
||
def cut4(inp):
|
||
inp = inp.strip("\n")
|
||
return "\n".join(["%s" % item for item in inp.strip(".").split(".")])
|
||
|
||
# 按标点符号切
|
||
# contributed by https://github.com/AI-Hobbyist/GPT-SoVITS/blob/main/GPT_SoVITS/inference_webui.py
|
||
@register_method("cut5")
|
||
def cut5(inp):
|
||
# if not re.search(r'[^\w\s]', inp[-1]):
|
||
# inp += '。'
|
||
inp = inp.strip("\n")
|
||
# punds = r'[,.;?!、，。？！;：…]'
|
||
punds = r'[,.;?!、，。？！；：:…]'
|
||
items = re.split(f'({punds})', inp)
|
||
mergeitems = ["".join(group) for group in zip(items[::2], items[1::2])]
|
||
# 在句子不存在符号或句尾无符号的时候保证文本完整
|
||
if len(items)%2 == 1:
|
||
mergeitems.append(items[-1])
|
||
opt = "\n".join(mergeitems)
|
||
return opt
|
||
|
||
def num_to_chinese(num):
|
||
chinese_nums = {
|
||
0: '零', 1: '一', 2: '二', 3: '三', 4: '四', 5: '五', 6: '六', 7: '七', 8: '八', 9: '九',
|
||
}
|
||
units = ['', '十', '百', '千', '万', '亿']
|
||
num_str = str(num)
|
||
num_str_rev = num_str[::-1]
|
||
result = ''
|
||
for i, digit in enumerate(num_str_rev):
|
||
if i == 0 and digit == '0':
|
||
continue
|
||
if i > 0 and digit == '0' and result[0] != '零':
|
||
result = '零' + result
|
||
digit_chinese = chinese_nums[int(digit)]
|
||
unit = units[i % 4]
|
||
if i % 4 == 0:
|
||
unit = units[i % 4 + int(i / 4)]
|
||
result = digit_chinese + unit + result
|
||
return result
|
||
|
||
# 支持语言混合切，按照约10个字一组，拆分更多的文本支持batch并行推理
|
||
@register_method("mixed_cut")
|
||
def mixed_cut(inp):
|
||
def re_exp_japenese_char():
|
||
#日文中带有中文字符的情况，依赖short合并把中文合并到上一个日文分组中
|
||
return '[\u3040-\u30FF\uFF66-\uFF9D]'
|
||
def re_exp_chinese_char():
|
||
return '[\u4e00-\u9fa5]'
|
||
def re_exp_alpha():
|
||
return '[a-zA-Z]'
|
||
def re_exp_digit():
|
||
return '[0-9]'
|
||
bad_case_ignore = [ "...","~","——","……" ]
|
||
for ss in bad_case_ignore:
|
||
inp = inp.replace(ss, "。")
|
||
result = []
|
||
last_s = ""
|
||
last_c_type = ""
|
||
#按连续字符进行分组
|
||
for char in inp:
|
||
c_type = "unknow"
|
||
if char == " ":
|
||
last_s += char
|
||
continue
|
||
elif re.match(re_exp_japenese_char(), char):
|
||
c_type="jps"
|
||
elif re.match(re_exp_chinese_char(), char):
|
||
c_type="hans"
|
||
elif re.match(re_exp_alpha(), char):
|
||
c_type="alpha"
|
||
elif re.match(re_exp_digit(), char):
|
||
c_type="digit"
|
||
if (c_type != last_c_type and c_type != "unknow" and len(last_c_type) > 0):
|
||
result.append(last_s)
|
||
last_s = ""
|
||
last_s += char
|
||
if c_type != "unknow":
|
||
last_c_type = c_type
|
||
elif len(last_s) > 10:
|
||
result.append(last_s)
|
||
last_s = ""
|
||
result.append(last_s)
|
||
|
||
def s_type(s):
|
||
if len(s) > 0:
|
||
if re.compile(re_exp_japenese_char()).search(s) is not None:
|
||
return "jps"
|
||
elif re.compile(re_exp_chinese_char()).search(s) is not None:
|
||
return "hans"
|
||
elif re.compile(re_exp_alpha()).search(s) is not None:
|
||
return "alpha"
|
||
elif re.compile(re_exp_digit()).search(s) is not None:
|
||
return "digit"
|
||
return "unknow"
|
||
#数组合并至前项，并支持念出中文数字
|
||
new_result = []
|
||
n = 0
|
||
while n < len(result):
|
||
this_s = result[n]
|
||
this_s_type = s_type(this_s)
|
||
before_s = ""
|
||
if n > 0:
|
||
before_s = result[n-1]
|
||
before_s_type = s_type(this_s)
|
||
next_s = ""
|
||
if n < (len(result)-1):
|
||
next_s = result[n+1]
|
||
next_s_type = s_type(this_s)
|
||
if this_s_type == "digit":
|
||
if before_s == "":
|
||
new_result.append(before_s)
|
||
if before_s_type == "hans" or next_s_type == "hans":
|
||
ss = num_to_chinese(this_s)
|
||
else:
|
||
ss = this_s
|
||
if before_s == "" or before_s_type == next_s_type:
|
||
ss += next_s
|
||
n+=1
|
||
new_result[len(new_result)-1]+=ss
|
||
else:
|
||
new_result.append(this_s)
|
||
n+=1
|
||
opt = "\n".join(new_result)
|
||
return opt
|
||
|
||
if __name__ == '__main__':
|
||
method = get_method("mixed_cut")
|
||
print(method("你好，我是小明。你好，我是小红。你好，我是小刚。你好，我是小张。") + "\n===\n")
|
||
print(method("你好，我是小明") + "\n===\n")
|
||
print(method("12345") + "\n===\n")
|
||
print(method("123，不许动") + "\n===\n")
|
||
print(method("你好，我是小明。我今年20岁了") + "\n===\n")
|
||
print(method("你好，我是Maxwell, nice to meet you") + "\n===\n")
|
||
print(method("你好，我是Maxwell。我今年20岁了") + "\n===\n")
|
||
print(method("你好，我是小明。こんにちは、シャオミンです。") + "\n===\n")
|
||
print(method("こんにちは、シャオミンです。 今年で20周年") + "\n===\n")
|
||
print(method("こんにちは、シャオミンです。 今年で20周年， nice to meet you") + "\n===\n")
|
||
print(method("こんにちは、シャオミンです。nice to meet you") + "\n===\n")
|
||
print(method("Hello, I am Maxwell. 20 years old，中文名叫小明") + "\n===\n")
|
||
|