Spaces:
Running
Running
File size: 2,151 Bytes
3dd84f8 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 |
import re
from text.english import english_to_ipa2
from text.mandarin import chinese_to_cnm3
from text.japanese import japanese_to_ipa2
language_module_map = {"PAD":0, "ZH": 1, "EN": 2, "JA": 3}
# 预编译正则表达式
ZH_PATTERN = re.compile(r'[\u3400-\u4DBF\u4e00-\u9FFF\uF900-\uFAFF\u3000-\u303F]')
EN_PATTERN = re.compile(r'[a-zA-Z.,!?\'"(){}[\]<>:;@#$%^&*-_+=/\\|~`]+')
JP_PATTERN = re.compile(r'[\u3040-\u309F\u30A0-\u30FF\u4E00-\u9FAF\u31F0-\u31FF\uFF00-\uFFEF\u3000-\u303F]')
CLEANER_PATTERN = re.compile(r'\[(ZH|EN|JA)\]')
def detect_language(text: str, prev_lang=None):
"""
根据给定的文本检测语言
:param text: 输入文本
:param prev_lang: 上一个检测到的语言
:return: 'ZH' for Chinese, 'EN' for English, 'JA' for Japanese, or prev_lang for spaces
"""
if ZH_PATTERN.search(text): return 'ZH'
if EN_PATTERN.search(text): return 'EN'
if JP_PATTERN.search(text): return 'JA'
if text.isspace(): return prev_lang # 若是空格，则返回前一个语言
return None
# auto detect language using re
def cjke_cleaners4(text: str):
"""
根据文本内容自动检测语言并转换为IPA音标
:param text: 输入文本
:return: 转换为IPA音标的文本
"""
text = CLEANER_PATTERN.sub('', text)
pointer = 0
output = ''
current_language = detect_language(text[pointer])
while pointer < len(text):
temp_text = ''
while pointer < len(text) and detect_language(text[pointer], current_language) == current_language:
temp_text += text[pointer]
pointer += 1
if current_language == 'ZH':
output += chinese_to_cnm3(temp_text)
elif current_language == 'JA':
output += japanese_to_ipa2(temp_text)
elif current_language == 'EN':
output += english_to_ipa2(temp_text)
if pointer < len(text):
current_language = detect_language(text[pointer])
output = re.sub(r'\s+$', '', output)
output = re.sub(r'([^\.,!\?\-…~])$', r'\1.', output)
return output
|