Kyrgyz CV dictionary v2.0.0#

Maintainer: Vox Communis
Language: Kyrgyz
Dialect: N/A
Phone set: Epitran
Number of words: 11,645
Phones: b d d͡ʒ e f i j k l m n o p r s t t͡s t͡ʃ u v y z ø ŋ ɑ ɡ ɯ ʃ
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}
}

If you have comments or questions about this dictionary or its phone set, you can check previous MFA model discussion posts or create a new one.

Acoustic models

Kyrgyz CV acoustic model v2.0.0

Installation#

Install from the MFA command line:

mfa model download dictionary kyrgyz_cv

Or download from the release page.

Intended use#

This dictionary is intended for forced alignment of Kyrgyz transcripts.

This dictionary uses the Epitran phone set for Kyrgyz, and was used in training the Kyrgyz 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	Alveopalatal	Palatal	Velar
Nasal	m Occurrences: 2,582 Examples: * адем: [ɑ d j e m] * азем: [ɑ z j e m] * кимди: [k i m d i] * бөлүм: [b ø l y m]		n Occurrences: 6,255 Examples: * сенин: [s j e n i n] * тобун: [t o b u n] * отун: [o t u n] * ничто: [n i t͡ʃ t o]			ŋ Occurrences: 307 Examples: * кожоң: [k o d͡ʒ o ŋ] * таңды: [t ɑ ŋ d ɯ] * өзүңө: [ø z y ŋ ø] * уңгу: [u ŋ ɡ u]
Stop	p Occurrences: 1,476 Examples: * кечпи: [k j e t͡ʃ p i] * ачып,: [ɑ t͡ʃ ɯ p] * кагып: [k ɑ ɡ ɯ p] * поль: [p o l] b Occurrences: 2,344 Examples: * тобун: [t o b u n] * бою: [b o j u] * бийик: [b i j i k] * күбө: [k y b ø]		t Occurrences: 5,981 Examples: * тобун: [t o b u n] * отун: [o t u n] * коет: [k o j e t] * дота: [d o t ɑ] d Occurrences: 3,785 Examples: * адем: [ɑ d j e m] * кимди: [k i m d i] * да.: [d ɑ] * дота: [d o t ɑ]			k Occurrences: 5,725 Examples: * кечпи: [k j e t͡ʃ p i] * кимди: [k i m d i] * бийик: [b i j i k] * кагып: [k ɑ ɡ ɯ p] ɡ Occurrences: 2,540 Examples: * сүлгү: [s y l ɡ y] * кагып: [k ɑ ɡ ɯ p] * го,: [ɡ o] * агама: [ɑ ɡ ɑ m ɑ]
Affricate			t͡s Occurrences: 187 Examples: * акциз: [ɑ k t͡s i z] * перцы: [p j e r t͡s ɯ] * цехти: [t͡s j e k t i] * хенце: [k j e n t͡s j e]	t͡ʃ Occurrences: 1,253 Examples: * кечпи: [k j e t͡ʃ p i] * ачып,: [ɑ t͡ʃ ɯ p] * ничто: [n i t͡ʃ t o] * чыкты: [t͡ʃ ɯ k t ɯ] d͡ʒ Occurrences: 1,246 Examples: * жалаа: [d͡ʒ ɑ l ɑ ɑ] * кожоң: [k o d͡ʒ o ŋ] * эжем,: [e d͡ʒ j e m] * жоюу: [d͡ʒ o j u u]
Sibilant			s Occurrences: 2,437 Examples: * сенин: [s j e n i n] * сүлгү: [s y l ɡ y] * сумма: [s u m m ɑ] * исак: [i s ɑ k] z Occurrences: 1,321 Examples: * азем: [ɑ z j e m] * зоок: [z o o k] * өзүм: [ø z y m] * өзүңө: [ø z y ŋ ø]	ʃ Occurrences: 1,777 Examples: * болуш: [b o l u ʃ] * ашуун: [ɑ ʃ u u n] * эшик: [e ʃ i k] * шыр: [ʃ ɯ r]
Fricative		f Occurrences: 176 Examples: * эфир: [e f i r] * фото: [f o t o] * кайф: [k ɑ j f] * форма: [f o r m ɑ] v Occurrences: 539 Examples: * акаев: [ɑ k ɑ j e v] * автор: [ɑ v t o r] * авто: [ɑ v t o] * завод: [z ɑ v o d]
Approximant					j Occurrences: 6,519 Examples: * сенин: [s j e n i n] * кечпи: [k j e t͡ʃ p i] * адем: [ɑ d j e m] * улай: [u l ɑ j]
Trill			r Occurrences: 5,260 Examples: * ырлар: [ɯ r l ɑ r] * эфир: [e f i r] * оор.: [o o r] * уран: [u r ɑ n]
Lateral			l Occurrences: 4,178 Examples: * улай: [u l ɑ j] * ырлар: [ɯ r l ɑ r] * сүлгү: [s y l ɡ y] * поль: [p o l]

Vowels#

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

	Front	Back
Close	i Occurrences: 4,228 Examples: * сенин: [s j e n i n] * кечпи: [k j e t͡ʃ p i] * кимди: [k i m d i] * бийик: [b i j i k] y Occurrences: 2,563 Examples: * сүлгү: [s y l ɡ y] * күбө: [k y b ø] * бөлүм: [b ø l y m] * мүлк: [m y l k]	ɯ Occurrences: 4,954 Examples: * ачып,: [ɑ t͡ʃ ɯ p] * ырлар: [ɯ r l ɑ r] * кагып: [k ɑ ɡ ɯ p] * чыкты: [t͡ʃ ɯ k t ɯ] u Occurrences: 4,104 Examples: * тобун: [t o b u n] * улай: [u l ɑ j] * бою: [b o j u] * отун: [o t u n]

Close-Mid	e Occurrences: 4,775 Examples: * сенин: [s j e n i n] * кечпи: [k j e t͡ʃ p i] * адем: [ɑ d j e m] * азем: [ɑ z j e m] ø Occurrences: 2,265 Examples: * күбө: [k y b ø] * бөлүм: [b ø l y m] * өчүк: [ø t͡ʃ y k] * көр.: [k ø r]	o Occurrences: 3,622 Examples: * тобун: [t o b u n] * бою: [b o j u] * отун: [o t u n] * поль: [p o l]

Open-Mid

Open		ɑ Occurrences: 11,831 Examples: * адем: [ɑ d j e m] * ачып,: [ɑ t͡ʃ ɯ p] * улай: [u l ɑ j] * азем: [ɑ z j e m]