jaconv-domino 0.1.0

Japanese character converter extending jaconv with pydomino phoneme conversion

jaconv-domino

jaconv-domino is a Japanese character converter library extending jaconv with pydomino phoneme conversion features.

日本語版 README

Installation

pip install jaconv-domino

About jaconv

For jaconv features, please refer to:

• PyPI: https://pypi.org/project/jaconv/
• GitHub: https://github.com/ikegami-yukino/jaconv

jaconv functions can be called as follows:

import jaconv_domino

# Hiragana to Katakana
jaconv_domino.hira2kata('ともえまみ')
# => 'トモエマミ'

# Katakana to Hiragana
jaconv_domino.kata2hira('マミサン')
# => 'まみさん'

# Hiragana to Half-width Katakana
jaconv_domino.hira2hkata('ともえまみ')
# => 'ﾄﾓｴﾏﾐ'

# Half-width to Full-width
jaconv_domino.h2z('ﾃｨﾛﾌｨﾅｰﾚ')
# => 'ティロフィナーレ'
jaconv_domino.hankaku2zenkaku('ﾃｨﾛﾌｨﾅｰﾚ')  # alias
jaconv_domino.han2zen('ﾃｨﾛﾌｨﾅｰﾚ')  # alias

# Full-width to Half-width
jaconv_domino.z2h('ティロフィナーレ')
# => 'ﾃｨﾛﾌｨﾅｰﾚ'
jaconv_domino.zenkaku2hankaku('ティロフィナーレ')  # alias
jaconv_domino.zen2han('ティロフィナーレ')  # alias

# Normalize (half-width kana to full-width, wave dash to long vowel mark, etc.)
jaconv_domino.normalize('ﾃｨﾛ･フィナ〜レ')
# => 'ティロ・フィナーレ'

# Kana to Romaji
jaconv_domino.kana2alphabet('ばなな')
# => 'banana'

# Romaji to Hiragana
jaconv_domino.alphabet2kana('ohayou')
# => 'おはよう'

# Katakana to Romaji
jaconv_domino.kata2alphabet('バナナ')
# => 'banana'

# Romaji to Katakana
jaconv_domino.alphabet2kata('ohayou')
# => 'オハヨウ'

# Small kana to normal kana
jaconv_domino.enlargesmallkana('ぁぃぅぇぉ')
# => 'あいうえお'

# Hiragana to Julius phoneme format
jaconv_domino.hiragana2julius('てんきすごくいいいいいい')
# => 't e N k i s u g o k u i:'

pydomino Phoneme Conversion (jaconv-domino specific)

Hiragana to Phoneme

import jaconv_domino

# Convert hiragana to pydomino phoneme format
jaconv_domino.hiragana2domino('ありがとう')
# => 'pau a ry i g a t o u pau'

# Newlines are converted to pau (consecutive pau markers are merged)
jaconv_domino.hiragana2domino('ありがとう\nこんにちは')
# => 'pau a ry i g a t o u pau k o N ny i ch i h a pau'

# Convert with character-phoneme mapping
phonemes, mapping = jaconv_domino.hiragana2domino_with_mapping('ありがとう')
# phonemes => 'pau a ry i g a t o u pau'
# mapping => [('あ', ['a']), ('り', ['ry', 'i']), ('が', ['g', 'a']), ('と', ['t', 'o']), ('う', ['u'])]

Phoneme to Hiragana

# Convert pydomino phoneme format to hiragana (pau is ignored)
jaconv_domino.domino2hiragana('pau a ry i g a t o u pau')
# => 'ありがとう'

jaconv_domino.domino2hiragana('k o N ny i ch i h a')
# => 'こんにちは'

# Convert phonemes with timing to hiragana with timing
phoneme_timings = [
    ("a", 0.0, 0.1),
    ("ry", 0.1, 0.15),
    ("i", 0.15, 0.2),
    ("g", 0.2, 0.25),
    ("a", 0.25, 0.3),
]
jaconv_domino.domino2hiragana_with_timing(phoneme_timings)
# => [('あ', 0.0, 0.1), ('り', 0.1, 0.2), ('が', 0.2, 0.3)]

CSV Format Conversion (for domino-song integration)

# CSV phoneme timing to hiragana string
# First and last pau are ignored, middle pau becomes newline
csv = """start,end,phoneme
0.000,0.620,pau
0.620,0.680,a
0.680,1.350,ry
1.350,2.010,i
2.010,2.100,pau
2.100,3.260,a
3.260,3.290,t
3.290,3.510,o
3.510,3.620,u
3.620,3.739,pau"""

jaconv_domino.domino_csv2hiragana(csv)
# => 'あり\nあとう'

# CSV phoneme timing to CSV hiragana timing
jaconv_domino.domino_csv2hiragana_csv(csv)
# => 'hiragana,start,end\nあ,0.62,0.68\nり,0.68,2.01\n...'

# Hiragana CSV to phoneme string
hiragana_csv = """あ,0.0,0.1
り,0.1,0.2
が,0.2,0.3
と,0.3,0.4
う,0.4,0.5"""

jaconv_domino.hiragana_csv2domino(hiragana_csv)
# => 'pau a ry i g a t o u pau'

Notes:

• Column order is auto-detected from header (start,end,phoneme or phoneme,start,end)
• pau (pause) phoneme: ignored in domino2*, newlines converted to pau in *2domino
• Characters other than hiragana, katakana, and alphabet (punctuation, symbols, long vowel marks, etc.) are ignored

API Reference

jaconv Re-exported Functions

Function	Description
hira2kata(text)	Hiragana to Full-width Katakana
hira2hkata(text)	Hiragana to Half-width Katakana
kata2hira(text)	Full-width Katakana to Hiragana
h2z(text)	Half-width to Full-width
z2h(text)	Full-width to Half-width
normalize(text)	Unicode normalization
kana2alphabet(text)	Kana to Romaji
alphabet2kana(text)	Romaji to Hiragana
kata2alphabet(text)	Katakana to Romaji
alphabet2kata(text)	Romaji to Katakana
enlargesmallkana(text)	Small kana to normal size
hiragana2julius(text)	Hiragana to Julius phoneme format

Aliases: hankaku2zenkaku=h2z, han2zen=h2z, zenkaku2hankaku=z2h, zen2han=z2h

jaconv-domino Specific Functions

Function	Description
hiragana2domino(text)	Hiragana to pydomino phoneme format
hiragana2domino_with_mapping(text)	Hiragana to phoneme (with character mapping)
domino2hiragana(phonemes)	pydomino phoneme to Hiragana
domino2hiragana_with_timing(timings)	Phoneme timing to Hiragana timing
domino_csv2hiragana(csv)	CSV phoneme timing to Hiragana string
domino_csv2hiragana_csv(csv)	CSV phoneme timing to CSV Hiragana timing
hiragana_csv2domino(csv)	Hiragana CSV to phoneme string

License

MIT License