Ctrl+K
Ctrl+K

Indonesian CV dictionary v2.0.0#

  • Maintainer: Vox Communis

  • Language: Indonesian

  • Dialect: N/A

  • Phone set: Epitran

  • Number of words: 14,928

  • Phones: a b d d͡ʑ f h i j k l m n o p r s t t͡ɕ u w x z ŋ ɕ ə ɡ ɲ ʔ

  • License: CC-0

  • Compatible MFA version: v2.0.0

  • Citation:

@misc{Ahn_Chodroff_2022,
	author={Ahn, Emily and Chodroff, Eleanor},
	title={VoxCommunis Corpus},
	address={\url{https://osf.io/t957v}},
	publisher={OSF},
	year={2022},
	month={Jan}
}

Installation#

Install from the MFA command line:

mfa model download dictionary indonesian_cv

Or download from the release page.

Intended use#

This dictionary is intended for forced alignment of Indonesian transcripts.

This dictionary uses the Epitran phone set for Indonesian, and was used in training the Indonesian Epitran acoustic model. Pronunciations can be added on top of the dictionary, as long as no additional phones are introduced.

Performance Factors#

When trying to get better alignment accuracy, adding pronunciations is generally helpful, especially for different styles and dialects. The most impactful improvements will generally be seen when adding reduced variants that involve deleting segments/syllables common in spontaneous speech. Alignment must include all phones specified in the pronunciation of a word, and each phone has a minimum duration (by default 10ms). If a speaker pronounces a multisyllabic word with just a single syllable, it can be hard for MFA to fit all the segments in, so it will lead to alignment errors on adjacent words as well.

Ethical considerations#

Deploying any Speech-to-Text model into any production setting has ethical implications. You should consider these implications before use.

Demographic Bias#

You should assume every machine learning model has demographic bias unless proven otherwise. For pronunciation dictionaries, it is often the case that transcription accuracy and lexicon coverage for the prestige variety modeled in this dictionary compared to other variants. If you are using this dictionary in production, you should acknowledge this as a potential issue.

IPA Charts#

Consonants#

Obstruent symbols to the left of are unvoiced and those to the right are voiced.

Manner

Labial

Labiodental

Alveolar

Palatal

Velar

Glottal

Nasal
Occurrences:
4,930
Examples:
* muka.:
[m u k a]
* lemot:
[l ə m o t]
* manis:
[m a n i s]
* umur:
[u m u r]
Occurrences:
6,993
Examples:
* not:
[n o t]
* duran:
[d u r a n]
* naik,:
[n a i k]
* rawan:
[r a w a n]
Occurrences:
1,560
Examples:
* bunny:
[b u n ɲ]
* nyeri:
[ɲ ə r i]
* pliny:
[p l i ɲ]
* nanya:
[n a ɲ a]
Occurrences:
2,085
Examples:
* sang:
[s a ŋ]
* gang,:
[ɡ a ŋ]
* hung:
[h u ŋ]
* angsa:
[a ŋ s a]

Stop
Occurrences:
3,272
Examples:
* pisau:
[p i s a u]
* lapuk:
[l a p u k]
* api,:
[a p i]
* sopan:
[s o p a n]
Occurrences:
2,984
Examples:
* baku:
[b a k u]
* biru?:
[b i r u]
* band:
[b a n d]
* barel:
[b a r ə l]
Occurrences:
5,487
Examples:
* not:
[n o t]
* lemot:
[l ə m o t]
* cinta:
[t͡ɕ i n t a]
* worst:
[w o r s t]
Occurrences:
2,917
Examples:
* duran:
[d u r a n]
* disco:
[d i s t͡ɕ o]
* band:
[b a n d]
* hood:
[h o o d]
Occurrences:
5,172
Examples:
* korea:
[k o r ə a]
* muka.:
[m u k a]
* baku:
[b a k u]
* naik,:
[n a i k]
Occurrences:
1,440
Examples:
* gila:
[ɡ i l a]
* garak:
[ɡ a r a k]
* regis:
[r ə ɡ i s]
* gen:
[ɡ ə n]
Occurrences:
46
Examples:
* 'kan?:
[ʔ k a n]
* 'kan:
[ʔ k a n]
* o'day:
[o ʔ d a j]

Affricate
Occurrences:
1,361
Examples:
* disco:
[d i s t͡ɕ o]
* rock.:
[r o t͡ɕ k]
* cinta:
[t͡ɕ i n t a]
* lucu.:
[l u t͡ɕ u]
Occurrences:
899
Examples:
* jas,:
[d͡ʑ a s]
* juli:
[d͡ʑ u l i]
* jet:
[d͡ʑ ə t]
* fuji,:
[f u d͡ʑ i]

Sibilant
Occurrences:
5,252
Examples:
* pisau:
[p i s a u]
* sang:
[s a ŋ]
* jas,:
[d͡ʑ a s]
* disco:
[d i s t͡ɕ o]
Occurrences:
135
Examples:
* pizza:
[p i z z a]
* hazel:
[h a z ə l]
* bezel:
[b ə z ə l]
* liz:
[l i z]
Occurrences:
26
Examples:
* syal:
[ɕ a l]
* tasya:
[t a ɕ a]

Fricative
Occurrences:
1,011
Examples:
* hafal:
[h a f a l]
* full:
[f u l l]
* fuji,:
[f u d͡ʑ i]
* fisik:
[f i s i k]
Occurrences:
2,389
Examples:
* hafal:
[h a f a l]
* marsh:
[m a r s h]
* thor.:
[t h o r]
* hazel:
[h a z ə l]

Approximant
Occurrences:
653
Examples:
* rawan:
[r a w a n]
* worst:
[w o r s t]
* kawan:
[k a w a n]
* sewa:
[s ə w a]
Occurrences:
645
Examples:
* kyoto:
[k j o t o]
* dusky:
[d u s k j]
* bayar:
[b a j a r]
* bavay:
[b a f a j]

Trill
Occurrences:
6,756
Examples:
* korea:
[k o r ə a]
* duran:
[d u r a n]
* biru?:
[b i r u]
* rawan:
[r a w a n]

Lateral
Occurrences:
4,306
Examples:
* lemot:
[l ə m o t]
* gila:
[ɡ i l a]
* lapuk:
[l a p u k]
* lulus:
[l u l u s]

Vowels#

Vowel symbols to the left of are unrounded and those to the right are rounded.

Front

Near-Front

Central

Near-Back

Back

Close
Occurrences:
8,316
Examples:
* gila:
[ɡ i l a]
* pisau:
[p i s a u]
* naik,:
[n a i k]
* biru?:
[b i r u]
Occurrences:
5,055
Examples:
* muka.:
[m u k a]
* duran:
[d u r a n]
* baku:
[b a k u]
* pisau:
[p i s a u]

Close-Mid
Occurrences:
3,756
Examples:
* korea:
[k o r ə a]
* not:
[n o t]
* lemot:
[l ə m o t]
* disco:
[d i s t͡ɕ o]
Occurrences:
9,701
Examples:
* korea:
[k o r ə a]
* lemot:
[l ə m o t]
* "more:
[m o r ə]
* jet:
[d͡ʑ ə t]

Open-Mid

Open
Occurrences:
17,034
Examples:
* korea:
[k o r ə a]
* muka.:
[m u k a]
* gila:
[ɡ i l a]
* duran:
[d u r a n]
On this page
Show Source