espnet2.text.phoneme_tokenizer.Jaso

source

class espnet2.text.phoneme_tokenizer.Jaso(space_symbol=' ', no_space=False)

Bases: object

A class for converting Korean text into Jamo characters.

This class takes Korean text as input and converts it into its corresponding Jamo characters. It also provides options to handle spaces in the output.

PUNC

A string of punctuation characters.

• Type: str

SPACE

A string representing a space character.

• Type: str

JAMO_LEADS

A string of Jamo leading characters.

• Type: str

JAMO_VOWELS

A string of Jamo vowel characters.

• Type: str

JAMO_TAILS

A string of Jamo tail characters.

• Type: str

VALID_CHARS

A string of valid characters, including Jamo characters, punctuation, and spaces.

• Type: str

• Parameters:

  • space_symbol (str) – The symbol to use for spaces in the output. Defaults to a regular space.
  • no_space (bool) – If True, spaces will be removed from the output.

Examples

>>> jaso = Jaso(space_symbol='<space>', no_space=False)
>>> jaso("안녕하세요")
['ᄋ', 'ᅡ', 'ᄂ', 'ᅣ', 'ᄉ', 'ᅥ', 'ᄒ', 'ᅡ', 'ᄋ', 'ᅭ']

>>> jaso_no_space = Jaso(no_space=True)
>>> jaso_no_space("안녕하세요")
['ᄋ', 'ᅡ', 'ᄂ', 'ᅣ', 'ᄉ', 'ᅥ', 'ᄒ', 'ᅡ', 'ᄋ', 'ᅭ']

JAMO_LEADS

*= 'ᄀᄁᄂᄃᄄᄅᄆᄇᄈᄉᄊᄋᄌᄍᄎᄏᄐᄑᄒ'*

JAMO_TAILS

*= 'ᆨᆩᆪᆫᆬᆭᆮᆯᆰᆱᆲᆳᆴᆵᆶᆷᆸᆹᆺᆻᆼᆽᆾᆿᇀᇁᇂ'*

JAMO_VOWELS

*= 'ᅡᅢᅣᅤᅥᅦᅧᅨᅩᅪᅫᅬᅭᅮᅯᅰᅱᅲᅳᅴᅵ'*

PUNC

*= "!'(),-.:;?"*

SPACE

*= ' '*

VALID_CHARS

*= "ᄀᄁᄂᄃᄄᄅᄆᄇᄈᄉᄊᄋᄌᄍᄎᄏᄐᄑ하ᅢᅣᅤᅥᅦᅧᅨᅩᅪᅫᅬᅭᅮᅯᅰᅱᅲᅳᅴᅵᆨᆩᆪᆫᆬᆭᆮᆯᆰᆱᆲᆳᆴᆵᆶᆷᆸᆹᆺᆻᆼᆽᆾᆿᇀᇁᇂ!'(),-.:;? "*