nicolaskodak / num2chinese

num2chinese

A nlp tool to transform numbers to Chinese

num2chinese uses regular expression to parse alphanumeric literals and transform them into readable Chinese charaters.

Why it matters

• Chinese's pronuncication has lots of exceptions.
• For Chinese numbers, a character is uttered dependent of context.
• Lots of rules are required to handle messy Chinese number pronunciation. Dont' reinvent the wheel!

Examples

• $120 : 美金一百二十
• 200塊 : 兩百塊
• 12121212個蘋果 : 一千兩百一十二萬一千兩百一十二個蘋果
• 2002002支 : 兩百萬兩千零二支
• 9487 : 九四八七
• 080080123 : 零八零零八零一二

Usage

text = '12121212個蘋果''
normalizer = Normalizer()
text_normalized = normalizer.normalize(text)
print(text_normalized)
# result is '一千兩百萬十二萬一千兩百一十二個蘋果'

Installation

pip install num2chinese

Requirements

python>=3.6,<4.0

License

MIT license