pengzhendong/g2p-mix

Grapheme-to-Phoneme for Mixed Chinese (Mandarin or Cantonese) and English

g2p-mix

• Cantonese: pycantonese
• English: g2p_en
• Mandarin: pypinyin

Usage

$ pip install g2p-mix
$ python

Mandarin

>>> from g2p_mix import G2pMix
>>> G2pMix().g2p("你这个idea, 不太make sense。", sandhi=True)

[
  { "word": "你", "phones": ["n", "i3"], "lang": "ZH" },
  { "word": "这", "phones": ["zh", "e4"], "lang": "ZH" },
  { "word": "个", "phones": ["g", "e4"], "lang": "ZH" },
  { "word": "idea", "phones": ["AY0", "D", "IY1", "AH0"], "lang": "EN" },
  { "word": ",", "phones": ",", "lang": "SYM" },
  { "word": "不", "phones": ["b", "u2"], "lang": "ZH" },
  { "word": "太", "phones": ["t", "ai4"], "lang": "ZH" },
  { "word": "make", "phones": ["M", "EY1", "K"], "lang": "EN" },
  { "word": "sense", "phones": ["S", "EH1", "N", "S"], "lang": "EN" },
  { "word": "。", "phones": "。", "lang": "SYM" }
]

Cantonese

>>> G2pMix(jyut=True).g2p("你这个idea, 不太make sense。")

[
  { "word": "你", "phones": ["n", "ei5"], "lang": "ZH" }
  { "word": "這", "phones": ["z", "e3"], "lang": "ZH" }
  { "word": "個", "phones": ["g", "o3"], "lang": "ZH" }
  { "word": "idea", "phones": ["AY0", "D", "IY1", "AH0"], "lang": "EN" }
  { "word": ",", "phones": ",", "lang": "SYM" }
  { "word": "不", "phones": ["b", "at1"], "lang": "ZH" }
  { "word": "太", "phones": ["t", "aai3"], "lang": "ZH" }
  { "word": "make", "phones": ["M", "EY1", "K"], "lang": "EN" }
  { "word": "sense", "phones": ["S", "EH1", "N", "S"], "lang": "EN" }
  { "word": "。", "phones": "。", "lang": "SYM" }
]

g2p-mix's Issues

致敬

大佬牛掰！解决了我现在面临的问题！本issue单纯致敬~