Korean Language

Download

This chapter describes the linguistic resources included in the file kor.zip of the "Ortolang repository".

List of phonemes

Consonant Plosives

SPPAS IPA Description Examples
b b voiced bilabial ㅂ, 보다
p p voiceless bilabial ㅂ
p_h pʰ voiceless bilabial aspirated ㅍ, 판다
p_> p͈ voiceless bilabial ejective ㅃ
t t voiceless alveolar ㄷ
t_h tʰ voiceless alveolar aspirated ㅌ
t_> t͈ voiceless alveolar ejective ㄸ, 때
d d voiced alveolar ㄷ, 달, 다
k k voiceless velar ㄱ, 목사로
k_h kʰ voiceless velar aspirated ㅋ
k_> k͈ voiceless velar ejective ㄲ, 껄껄
g g voiced velar ㄱ, 개, 곧

Consonant Fricatives

SPPAS IPA Description Examples
s s voiceless alveolar 산, 수, 싶어
s_> s͈ voiceless alveolar ejective ㅆ

Consonant Nasals

SPPAS IPA Description Examples
m m bilabial ㅁ, 및 , 모자
n n alveolar ㄴ, 보낸
N ŋ voiced velar 종이, 가방, 학생

Consonant Liquids

SPPAS IPA Description Examples
l l alveolar lateral ㄹ, 를, 술
4 ɾ alveolar flap ㄹ, 갈래, 날이

Semivowels

SPPAS IPA Description Examples
j j voiced palatal ㅛ, ㅠ, ㅑ, ㅒ, 예
w w voiced labiovelar ㅟ, ㅞ, ㅙ, ㅘ, ㅝ

Vowels

SPPAS IPA Description Examples
E ɛ open-mid front unrounded ㅒ, 얘
A ɑ open back unrounded ㅑ, 야
i i close front unrounded ㅣ, 이
e e close-mid front unrounded ㅔ, 에
o o close-mid back rounded ㅗ, 오
u u close back rounded ㅜ, 우
2 ø close-mid front rounded ㅚ
V ʌ open-mid back unrounded ㅓ, 어
M ɯ close back unrounded ㅡ, 으

Affricates

SPPAS IPA Description Examples
dz d͡z voiced alveolar 주
dZ d͡ʒ voiced postalveolar ㅈ
tS_> t͡ʃ͈ voiceless postalveolar ejective ㅊ
tS_h t͡ʃʰ voiceless postalveolar aspirated ㅉ

Fillers

SPPAS Description
laugh laughter
noise noises, unintelligible speech
dummy un-transcribed speech

Lexicons

All lexicons are (c) Laboratoire Parole et Langage, Aix-en-Provence, France:

  • kor.vocab contains a list of 33k different words;
  • kor.repl allows to convert symbols and abbreviations into a text form.

All of them are distributed under the terms of the GNU General Public License.

Help is needed to create the file kor_num.repl allowing SPPAS to convert numbers to their written form

Pronunciation dictionary

The Korean pronunciation dictionary was manually created and is still under construction. Any help is welcome!

It is distributed under the terms of the GNU General Public License.

Acoustic Model

The acoustic model was NOT trained from data. Monophones of other models were cut and pasted to create this one, mainly the English and Taiwanese models.

Korean data are welcome! Because data implies a better acoustic model then better alignments…

It is distributed under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International Public License.

The model was created using a Python script available in the SPPAS package: acmtrain.py.