Files
tts_site/vendor/melo/text/cleaner_multiling.py
claude f197177ae1 한국어 우선 MeloTTS 웹 서비스 추가 (GPU, 속도/피치, 로그인 없음)
- FastAPI 백엔드: /api/voices, /api/tts, /api/health
- 한국어 메인 + 영어(5종 악센트), 중국어/일본어 UI 제거
- 언어별 목소리 모델 선택, 속도(0.5~2.0x)/피치(±12반음) 조절
- 로그인/사용제한/과금 유도 없음 (무제한 테스트)
- NVIDIA GPU 자동 사용 (torch cu128, Blackwell sm_120)
- 유저 친화 한국어 UI (frontend)
- Docker/compose (GPU 예약), 8788 포트
- MeloTTS(순수 파이썬) 벤더링 + 검증된 버전 고정(constraints)
2026-08-23 21:57:48 +09:00

111 lines
2.6 KiB
Python
Raw Permalink Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
"""Set of default text cleaners"""
# TODO: pick the cleaner for languages dynamically
import re
# Regular expression matching whitespace:
_whitespace_re = re.compile(r"\s+")
rep_map = {
":": ",",
";": ",",
",": ",",
"。": ".",
"!": "!",
"?": "?",
"\n": ".",
"·": ",",
"、": ",",
"...": ".",
"…": ".",
"$": ".",
"“": "'",
"”": "'",
"‘": "'",
"’": "'",
"(": "'",
")": "'",
"(": "'",
")": "'",
"《": "'",
"》": "'",
"【": "'",
"】": "'",
"[": "'",
"]": "'",
"—": "",
"~": "-",
"~": "-",
"「": "'",
"」": "'",
}
def replace_punctuation(text):
pattern = re.compile("|".join(re.escape(p) for p in rep_map.keys()))
replaced_text = pattern.sub(lambda x: rep_map[x.group()], text)
return replaced_text
def lowercase(text):
return text.lower()
def collapse_whitespace(text):
return re.sub(_whitespace_re, " ", text).strip()
def remove_punctuation_at_begin(text):
return re.sub(r'^[,.!?]+', '', text)
def remove_aux_symbols(text):
text = re.sub(r"[\<\>\(\)\[\]\"\«\»\']+", "", text)
return text
def replace_symbols(text, lang="en"):
"""Replace symbols based on the lenguage tag.
Args:
text:
Input text.
lang:
Lenguage identifier. ex: "en", "fr", "pt", "ca".
Returns:
The modified text
example:
input args:
text: "si l'avi cau, diguem-ho"
lang: "ca"
Output:
text: "si lavi cau, diguemho"
"""
text = text.replace(";", ",")
text = text.replace("-", " ") if lang != "ca" else text.replace("-", "")
text = text.replace(":", ",")
if lang == "en":
text = text.replace("&", " and ")
elif lang == "fr":
text = text.replace("&", " et ")
elif lang == "pt":
text = text.replace("&", " e ")
elif lang == "ca":
text = text.replace("&", " i ")
text = text.replace("'", "")
elif lang== "es":
text=text.replace("&","y")
text = text.replace("'", "")
return text
def unicleaners(text, cased=False, lang='en'):
"""Basic pipeline for Portuguese text. There is no need to expand abbreviation and
numbers, phonemizer already does that"""
if not cased:
text = lowercase(text)
text = replace_punctuation(text)
text = replace_symbols(text, lang=lang)
text = remove_aux_symbols(text)
text = remove_punctuation_at_begin(text)
text = collapse_whitespace(text)
text = re.sub(r'([^\.,!\?\-…])$', r'\1.', text)
return text