tuna2134/sbv2-v2
1
0
Fork 0
Code Issues Pull Requests 1 Actions Packages Projects Releases Wiki Activity

Refactor: split style_bert_vits2.nlp.chinese package

Browse Source 
Configured so that the same public function is exported from the module with the same name for each language.
This commit is contained in:
tsukumi
2024-03-08 09:04:40 +00:00
parent a672aeefd9
commit df68771651
6 changed files with 210 additions and 206 deletions
Download Patch File Download Diff File Expand all files Collapse all files

24
style_bert_vits2/nlp/english/normalizer.py
View File

@@ -14,22 +14,12 @@ __NUMBER_PATTERN = re.compile(r"[0-9]+")
def normalize_text(text: str) -> str:
text = __normalize_numbers(text)
text = __replace_punctuation(text)
text = replace_punctuation(text)
text = re.sub(r"([,;.\?\!])([\w])", r"\1 \2", text)
return text
def __normalize_numbers(text: str) -> str:
text = re.sub(__COMMA_NUMBER_PATTERN, __remove_commas, text)
text = re.sub(__POUNDS_PATTERN, r"\1 pounds", text)
text = re.sub(__DOLLARS_PATTERN, __expand_dollars, text)
text = re.sub(__DECIMAL_NUMBER_PATTERN, __expand_decimal_point, text)
text = re.sub(__ORDINAL_PATTERN, __expand_ordinal, text)
text = re.sub(__NUMBER_PATTERN, __expand_number, text)
return text
def __replace_punctuation(text: str) -> str:
def replace_punctuation(text: str) -> str:
REPLACE_MAP = {
"：": ",",
"；": ",",
@@ -80,6 +70,16 @@ def __replace_punctuation(text: str) -> str:
return replaced_text
def __normalize_numbers(text: str) -> str:
text = re.sub(__COMMA_NUMBER_PATTERN, __remove_commas, text)
text = re.sub(__POUNDS_PATTERN, r"\1 pounds", text)
text = re.sub(__DOLLARS_PATTERN, __expand_dollars, text)
text = re.sub(__DECIMAL_NUMBER_PATTERN, __expand_decimal_point, text)
text = re.sub(__ORDINAL_PATTERN, __expand_ordinal, text)
text = re.sub(__NUMBER_PATTERN, __expand_number, text)
return text
def __expand_dollars(m: re.Match[str]) -> str:
match = m.group(1)
parts = match.split(".")