|
|
@ -5,13 +5,11 @@ import re
|
|
|
|
import cn2an
|
|
|
|
import cn2an
|
|
|
|
from pypinyin import lazy_pinyin, Style
|
|
|
|
from pypinyin import lazy_pinyin, Style
|
|
|
|
|
|
|
|
|
|
|
|
import sys
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
sys.path.append("/data/docker/liujing04/gpt-vits/gpt-vits-master")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
from text.symbols import punctuation
|
|
|
|
from text.symbols import punctuation
|
|
|
|
from text.tone_sandhi import ToneSandhi
|
|
|
|
from text.tone_sandhi import ToneSandhi
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
normalizer = lambda x: cn2an.transform(x, "an2cn")
|
|
|
|
|
|
|
|
|
|
|
|
current_file_path = os.path.dirname(__file__)
|
|
|
|
current_file_path = os.path.dirname(__file__)
|
|
|
|
pinyin_to_symbol_map = {
|
|
|
|
pinyin_to_symbol_map = {
|
|
|
|
line.split("\t")[0]: line.strip().split("\t")[1]
|
|
|
|
line.split("\t")[0]: line.strip().split("\t")[1]
|
|
|
@ -151,10 +149,8 @@ def _g2p(segments):
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
def text_normalize(text):
|
|
|
|
def text_normalize(text):
|
|
|
|
numbers = re.findall(r"\d+(?:\.?\d+)?", text)
|
|
|
|
dest_text=normalizer(text)
|
|
|
|
for number in numbers:
|
|
|
|
text = replace_punctuation(dest_text)
|
|
|
|
text = text.replace(number, cn2an.an2cn(number), 1)
|
|
|
|
|
|
|
|
text = replace_punctuation(text)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
return text
|
|
|
|
return text
|
|
|
|
|
|
|
|
|
|
|
|