from tensorflow_tts.processor.ch_pinyin import PINYIN_DICT, mark, zh_pattern
_pad = ["pad"]
_eos = ["eos"]
_pause = ["sil", "#0", "#1", "#2", "#3"]
# 声母
_initials = ["^", "b", "c", "ch", "d", "f", "g", "h", "j", "k", "l", "m", "n", "p", "q", "r", "s", "sh", "t", "x", "z",
"zh"]
_tones = ["1", "2", "3", "4", "5"]
# 韵母
_finals = ["a", "ai", "an", "ang", "ao", "e", "ei", "en", "eng", "er", "i", "ia", "ian", "iang", "iao", "ie", "ii",
"iii", "in", "ing", "iong", "iou", "o", "ong", "ou", "u", "ua", "uai", "uan", "uang", "uei", "uen", "ueng",
"uo", "v", "van", "ve", "vn"]
g2p = grapheme_to_phonem.G2p()
valid_symbols = g2p.phonemes
_punctuation = "!'(),.:;? "
_arpabet = ["@" + s for s in valid_symbols] + list(_punctuation)
BAKER_SYMBOLS = _pad + _pause + _initials + _finals + ["@" + i for i in _tones] + _arpabet + _eos
# BAKER_SYMBOLS = _pad + _pause + _initials + [i + j for i in _finals for j in _tones] + _arpabet + _eos
def is_zh(word):
match = zh_pattern.search(word)
return match is not None
print("aaa")
labels = np.array([[1, 3, 5], [2, 4, 6], [0, 0, 0]])
print(labels[None, ...])
print(labels[None, :])
_whitespace_re = re.compile(r"\s+")
print(re.sub(_whitespace_re, " ", "nihao ma wo hen hao"))
text = "nihaoma wo {B IY}henhao a "
print(len(text))
_curly_re = re.compile(r"(.*?)\{(.+?)\}(.*)")
print(_curly_re.match(text))
from g2p_en import g2p
g2p = g2p.G2p()
print(g2p('Ｂ'))
# print("==========="*2)
test = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz"
for i in test:
print(g2p(i))
print(g2p('tv'))
print(g2p('ad'))
# print(g2p('fetch'))
# print(g2p('cake'))
# print(g2p('age'))
# print(g2p('banana'))
# print(g2p('ABCDEFG'))
#
# texts = ""