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

Fmt only (maybe)

Browse Source
This commit is contained in:
litagin02
2024-05-25 18:15:36 +09:00
parent acf93b80ce
commit 2274087da4
33 changed files with 216 additions and 166 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
style_bert_vits2/nlp/chinese/g2p.py
View File

@@ -8,7 +8,7 @@ from style_bert_vits2.nlp.chinese.tone_sandhi import ToneSandhi
from style_bert_vits2.nlp.symbols import PUNCTUATIONS
with open(Path(__file__).parent / "opencpop-strict.txt", "r", encoding="utf-8") as f:
with open(Path(__file__).parent / "opencpop-strict.txt", encoding="utf-8") as f:
__PINYIN_TO_SYMBOL_MAP = {
line.split("\t")[0]: line.strip().split("\t")[1] for line in f.readlines()
}
@@ -73,7 +73,7 @@ def __g2p(segments: list[str]) -> tuple[list[str], list[int], list[int]]:
"iou": "iu",
"uen": "un",
}
if v_without_tone in v_rep_map.keys():
if v_without_tone in v_rep_map:
pinyin = c + v_rep_map[v_without_tone]
else:
# 单音节
@@ -83,7 +83,7 @@ def __g2p(segments: list[str]) -> tuple[list[str], list[int], list[int]]:
"in": "yin",
"u": "wu",
}
if pinyin in pinyin_rep_map.keys():
if pinyin in pinyin_rep_map:
pinyin = pinyin_rep_map[pinyin]
else:
single_rep_map = {
@@ -92,10 +92,10 @@ def __g2p(segments: list[str]) -> tuple[list[str], list[int], list[int]]:
"i": "y",
"u": "w",
}
if pinyin[0] in single_rep_map.keys():
if pinyin[0] in single_rep_map:
pinyin = single_rep_map[pinyin[0]] + pinyin[1:]
assert pinyin in __PINYIN_TO_SYMBOL_MAP.keys(), (
assert pinyin in __PINYIN_TO_SYMBOL_MAP, (
pinyin,
seg,
raw_pinyin,

2
style_bert_vits2/nlp/chinese/normalizer.py
View File

@@ -51,7 +51,7 @@ def normalize_text(text: str) -> str:
def replace_punctuation(text: str) -> str:
text = text.replace("嗯", "恩").replace("呣", "母")
pattern = re.compile("|".join(re.escape(p) for p in __REPLACE_MAP.keys()))
pattern = re.compile("|".join(re.escape(p) for p in __REPLACE_MAP))
replaced_text = pattern.sub(lambda x: __REPLACE_MAP[x.group()], text)

54
style_bert_vits2/nlp/chinese/tone_sandhi.py
View File

@@ -471,26 +471,27 @@ class ToneSandhi:
):
finals[j] = finals[j][:-1] + "5"
ge_idx = word.find("个")
if len(word) >= 1 and word[-1] in "吧呢啊呐噻嘛吖嗨呐哦哒额滴哩哟喽啰耶喔诶":
finals[-1] = finals[-1][:-1] + "5"
elif len(word) >= 1 and word[-1] in "的地得":
finals[-1] = finals[-1][:-1] + "5"
# e.g. 走了, 看着, 去过
# elif len(word) == 1 and word in "了着过" and pos in {"ul", "uz", "ug"}:
# finals[-1] = finals[-1][:-1] + "5"
elif (
len(word) > 1
and word[-1] in "们子"
and pos in {"r", "n"}
and word not in self.must_not_neural_tone_words
if (
len(word) >= 1
and word[-1] in "吧呢啊呐噻嘛吖嗨呐哦哒额滴哩哟喽啰耶喔诶"
or len(word) >= 1
and word[-1] in "的地得"
or (
(
len(word) > 1
and word[-1] in "们子"
and pos in {"r", "n"}
and word not in self.must_not_neural_tone_words
)
or len(word) > 1
and word[-1] in "上下里"
and pos in {"s", "l", "f"}
)
or len(word) > 1
and word[-1] in "来去"
and word[-2] in "上下进出回过起开"
):
finals[-1] = finals[-1][:-1] + "5"
# e.g. 桌上, 地下, 家里
elif len(word) > 1 and word[-1] in "上下里" and pos in {"s", "l", "f"}:
finals[-1] = finals[-1][:-1] + "5"
# e.g. 上来, 下去
elif len(word) > 1 and word[-1] in "来去" and word[-2] in "上下进出回过起开":
finals[-1] = finals[-1][:-1] + "5"
# 个做量词
elif (
ge_idx >= 1
@@ -500,12 +501,11 @@ class ToneSandhi:
)
) or word == "个":
finals[ge_idx] = finals[ge_idx][:-1] + "5"
else:
if (
word in self.must_neural_tone_words
or word[-2:] in self.must_neural_tone_words
):
finals[-1] = finals[-1][:-1] + "5"
elif (
word in self.must_neural_tone_words
or word[-2:] in self.must_neural_tone_words
):
finals[-1] = finals[-1][:-1] + "5"
word_list = self._split_word(word)
finals_list = [finals[: len(word_list[0])], finals[len(word_list[0]) :]]
@@ -549,10 +549,8 @@ class ToneSandhi:
if finals[i + 1][-1] == "4":
finals[i] = finals[i][:-1] + "2"
# "一" before non-tone4 should be yi4, e.g. 一天
else:
# "一" 后面如果是标点，还读一声
if word[i + 1] not in self.punc:
finals[i] = finals[i][:-1] + "4"
elif word[i + 1] not in self.punc:
finals[i] = finals[i][:-1] + "4"
return finals
def _split_word(self, word: str) -> list[str]: