Ukrainian MFA dictionary v2.0.0#
@techreport{mfa_ukrainian_mfa_dictionary_2022,
author={McAuliffe, Michael and Sonderegger, Morgan},
title={Ukrainian MFA dictionary v2.0.0},
address={\url{https://mfa-models.readthedocs.io/pronunciation dictionary/Ukrainian/Ukrainian MFA dictionary v2_0_0.html}},
year={2022},
month={Mar},
}
G2P models Acoustic models |
Installation#
Install from the MFA command line:
mfa model download dictionary ukrainian_mfa
Or download from the release page.
The dictionary available from the release page and command line installation has pronunciation and silence probabilities estimated as part acoustic model training (see Silence probability format and training pronunciation probabilities for more information. If you would like to use the version of this dictionary without probabilities, please see the plain dictionary.
Intended use#
This dictionary is intended for forced alignment of Ukrainian transcripts.
This dictionary uses the MFA phone set for Ukrainian, and was used in training the Ukrainian MFA 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 |
Dental |
Alveolar |
Alveopalatal |
Palatal |
Velar |
Glottal |
---|---|---|---|---|---|---|---|---|
Nasal |
Occurrences: 14,009 Examples: * томас: [t̪ ɔ m ɑ s̪] * нум: [n̪ ʊ m] * ймемо: [i m e m ɔ] * мойри: [m ɔ j ɾ ɪ] Occurrences: 1,484 Examples: * нгамі: [n̪ ɦ ɑ mʲ i] * міг: [mʲ i ɦ] * умій: [ʊ mʲ i i] * тиміш: [t̪ e mʲ i ʃ] Occurrences: 12 Examples: |
Occurrences: 20,986 Examples: * нічне: [ɲ i tʃ n̪ e] * нгамі: [n̪ ɦ ɑ mʲ i] * налию: [n̪ ɐ l ɪ j ʊ] * зночі: [z̪ n̪ ɔ tʃʲ i] Occurrences: 443 Examples: * ванна: [ʋ ɑ n̪ː ɐ] * цінну: [tsʲ i n̪ː ʊ] * панну: [p ɑ n̪ː ʊ] * кінну: [c i n̪ː ʊ] |
Occurrences: 6,035 Examples: * нічне: [ɲ i tʃ n̪ e] * їхній: [j i x ɲ i i] * ніц: [ɲ i t̪s̪] * давні: [d̪ ɑ u ɲ i] Occurrences: 1,199 Examples: * винні: [ʋ ɪ ɲː i] * рання: [ɾ ɑ ɲː ɐ] * мення: [m e ɲː ɐ] * вання: [ʋ ɑ ɲː ɐ] |
|||||
Stop |
Occurrences: 15,797 Examples: * плоха: [p l ɔ x ɐ] * полою: [p ɔ l ɔ j ʊ] * плила: [p l ɪ l ɐ] * п'єте: [p j ɛ t̪ e] Occurrences: 7,063 Examples: * буває: [b ʊ ʋ ɑ j e] * зруб: [z̪ ɾ u b] * богам: [b ɔ ɦ ɐ m] * бити: [b ɪ t̪ ɪ] Occurrences: 3 Examples: |
Occurrences: 18,356 Examples: * томас: [t̪ ɔ m ɑ s̪] * нести: [n̪ e s̪ t̪ ɪ] * круту: [k ɾ ʊ t̪ ʊ] * п'єте: [p j ɛ t̪ e] Occurrences: 20 Examples: * гетто: [ɦ ɛ t̪ː ɔ] Occurrences: 11,380 Examples: * ззаду: [z̪ː ɑ d̪ ʊ] * давні: [d̪ ɑ u ɲ i] * надаю: [n̪ ɐ d̪ ɐ j ʊ] * дивом: [d̪ e ʋ ɔ m] Occurrences: 120 Examples: * будда: [b u d̪ː ɐ] * оддає: [ɔ d̪ː ɑ j e] * оддам: [ɔ d̪ː ɐ m] * міддю: [mʲ i d̪ː ʊ] |
Occurrences: 1,147 Examples: * кітці: [c i ɔ tsʲː i] * кішку: [c i ʃ k ʊ] * луків: [l ʊ c i u] * шкіру: [ʃ c i ɾ ʊ] Occurrences: 1 Examples: |
Occurrences: 17,650 Examples: * який: [j ɐ k ɪ i] * круту: [k ɾ ʊ t̪ ʊ] * синку: [s̪ ɪ n̪ k ʊ] * отрок: [ɔ t̪ ɾ ɔ k] Occurrences: 8 Examples: * мекку: [m ɛ kː ʊ] * мекка: [m ɛ kː ɐ] Occurrences: 133 Examples: * аякже: [ɐ ɐ ɡ ʒ ɛ] * ґатов: [ɡ ɐ t̪ ɔ u] * ґміни: [ɡ mʲ i n̪ ɪ] * ґанок: [ɡ ɑ n̪ ɔ k] Occurrences: 1 Examples: * меґґі: [m e ɡː i] |
||||
Affricate |
Occurrences: 934 Examples: * ніц: [ɲ i t̪s̪] * цехом: [t̪s̪ ɛ x ɔ m] * оце: [ɔ t̪s̪ ɛ] * цезар: [t̪s̪ ɛ z̪ ɑ ɾ] Occurrences: 4 Examples: * цска: [t̪s̪ː k ɐ] Occurrences: 412 Examples: * гудзя: [ɦ u d̪z̪ ɐ] * гудзь: [ɦ u d̪z̪] * будз: [b u d̪z̪] * дзень: [d̪z̪ ɛ ɲ] Occurrences: 1 Examples: |
Occurrences: 7,505 Examples: * нічне: [ɲ i tʃ n̪ e] * учора: [ʊ tʃ ɔ ɾ ɐ] * очам: [ɔ tʃ ɑ m] * хащах: [x ɑ ʃ tʃ ɑ x] Occurrences: 95 Examples: * лучче: [l u tʃː e] * матч: [m ɐ tʃː] * одчай: [ɔ tʃː ɐ i] * одчув: [ɔ tʃː u u] Occurrences: 571 Examples: * джері: [dʒ ɛ ɾʲ i] * джеря: [dʒ e ɾ ɐ] * ходжу: [x ɔ dʒ ʊ] * воджу: [ʋ ɔ dʒ ʊ] |
||||||
Sibilant |
Occurrences: 13,344 Examples: * гусар: [ɦ u s̪ ɐ ɾ] * офіс: [ɔ fʲ i s̪] * томас: [t̪ ɔ m ɑ s̪] * схилі: [s̪ x ɪ ʎ i] Occurrences: 25 Examples: * ссав: [s̪ː ɐ u] * ссе: [s̪ː e] * масса: [m ɐ s̪ː ɐ] * ссати: [s̪ː ɑ t̪ ɪ] Occurrences: 9,595 Examples: * захар: [z̪ ɐ x ɑ ɾ] * зночі: [z̪ n̪ ɔ tʃʲ i] * зруб: [z̪ ɾ u b] * алмаз: [ɐ lː m ɑ z̪] Occurrences: 25 Examples: * ззаду: [z̪ː ɑ d̪ ʊ] * ззаді: [z̪ː ɑ dʲ i] |
Occurrences: 10,399 Examples: * місію: [mʲ i sʲ i j ʊ] * асія: [ɐ sʲ i j ɐ] * стій: [sʲ tʲ i i] * сіли: [sʲ i l ɪ] Occurrences: 108 Examples: * мессі: [m e sʲː i] * россю: [ɾ o sʲː u] * отся: [ɔ sʲː ɐ] * отсі: [ɔ sʲː i] Occurrences: 1,529 Examples: * злі: [zʲ ʎ i] * зніме: [zʲ ɲ i m ɛ] * возі: [ʋ ɔ zʲ i] * змію: [zʲ mʲ i j ʊ] Occurrences: 8 Examples: |
Occurrences: 5,878 Examples: * рвеш: [ɾ ʋ ɛ ʃ] * хащах: [x ɑ ʃ tʃ ɑ x] * щоки: [ʃ tʃ ɔ k ɪ] * прощу: [p ɾ ɔ ʃ tʃ ʊ] Occurrences: 150 Examples: * груші: [ɦ ɾ u ʃʲ i] * суші: [s̪ u ʃʲ i] * парші: [p ɑ ɾ ʃʲ i] * шіня: [ʃʲ i ɲ ɐ] Occurrences: 3 Examples: Occurrences: 3,249 Examples: * вжив: [ʋ ʒ ɪ u] * жуков: [ʒ u k ɔ u] * тож: [t̪ ɔ ʒ] * жнива: [ʒ n̪ ɪ ʋ ɐ] Occurrences: 68 Examples: * етажі: [e t̪ ɐ ʒʲ i] * жін: [ʒʲ i n̪] * жінка: [ʒʲ i n̪ k ɐ] * жінко: [ʒʲ i n̪ k ɔ] Occurrences: 23 Examples: |
|||||
Fricative |
Occurrences: 827 Examples: * формі: [f ɔ ɾ m i] * ферм: [f ɛ ɾ m] * айфон: [ɐ i f ɔ n̪] * рифи: [ɾ ɪ f ɪ] Occurrences: 334 Examples: * офіс: [ɔ fʲ i s̪] * офісу: [ɔ fʲ i s̪ ʊ] * шафі: [ʃ ɑ fʲ i] * фішка: [fʲ i ʃ k ɐ] |
Occurrences: 232 Examples: * хівря: [ç i u ɾʲ ɐ] * тихім: [t̪ ɪ ç i m] * духів: [d̪ ʊ ç i u] * вхіду: [ʋ ç i d̪ ʊ] Occurrences: 514 Examples: * гірко: [ʝ i ɾ k ɔ] * гірка: [ʝ i ɾ k ɐ] * гімн: [ʝ i m n̪] * гіфи: [ʝ i f ɪ] |
Occurrences: 8,196 Examples: * гусар: [ɦ u s̪ ɐ ɾ] * нгамі: [n̪ ɦ ɑ mʲ i] * міг: [mʲ i ɦ] * богам: [b ɔ ɦ ɐ m] Occurrences: 5 Examples: * реггі: [ɾ e ɦː i] |
|||||
Approximant |
Occurrences: 16,468 Examples: * буває: [b ʊ ʋ ɑ j e] * слову: [s̪ l ɔ ʋ ʊ] * вжив: [ʋ ʒ ɪ u] * рвеш: [ɾ ʋ ɛ ʃ] Occurrences: 3,568 Examples: * квіти: [k ʋʲ i t̪ ɪ] * ловів: [l ɔ ʋʲ i u] * вівса: [ʋʲ i u s̪ ɐ] * відти: [ʋʲ i d̪ t̪ ɪ] Occurrences: 21 Examples: * ввів: [ʋʲː i u] * вві: [ʋʲː i] Occurrences: 46 Examples: * ввдно: [ʋː d̪ n̪ ɔ] * вволю: [ʋː ɔ ʎ ʊ] * ввело: [ʋː e l ɔ] * ввесь: [ʋː ɛ sʲ] |
Occurrences: 9,709 Examples: * буває: [b ʊ ʋ ɑ j e] * діяв: [dʲ i j ɑ u] * налию: [n̪ ɐ l ɪ j ʊ] * мойри: [m ɔ j ɾ ɪ] |
||||||
Tap |
Occurrences: 23,819 Examples: * гусар: [ɦ u s̪ ɐ ɾ] * захар: [z̪ ɐ x ɑ ɾ] * учора: [ʊ tʃ ɔ ɾ ɐ] * мойри: [m ɔ j ɾ ɪ] Occurrences: 2,795 Examples: * горює: [ɦ ɔ ɾʲ ʊ e] * стрій: [s̪ tʲ ɾʲ i i] * нарік: [n̪ ɑ ɾʲ i k] * рід: [ɾʲ i d̪] Occurrences: 6 Examples: * гаррі: [ɦ ɐ ɾʲː i] Occurrences: 14 Examples: * ферро: [f ɛ ɾː ɔ] * гурра: [ɦ ʊ ɾː ɐ] |
|||||||
Lateral |
Occurrences: 15,940 Examples: * налию: [n̪ ɐ l ɪ j ʊ] * плоха: [p l ɔ x ɐ] * слову: [s̪ l ɔ ʋ ʊ] * мало: [m ɑ l ɔ] Occurrences: 15 Examples: * алмаз: [ɐ lː m ɑ z̪] * алвіш: [ɐ lː ʋʲ i ʃ] * алло: [ɐ lː ɔ] * аллах: [ɐ lː ɑ x] |
Occurrences: 5,939 Examples: * схилі: [s̪ x ɪ ʎ i] * злі: [zʲ ʎ i] * конлі: [k ɔ ɲ ʎ i] * людей: [ʎ u d̪ ɛ i] Occurrences: 90 Examples: * валль: [ʋ ɐ ʎː] * виллє: [ʋ ɪ ʎː ɛ] * ілліч: [i ʎː i tʃ] * гіллі: [ʝ i ʎː i] |
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: 31,971 Examples: * нічне: [ɲ i tʃ n̪ e] * офіс: [ɔ fʲ i s̪] * нгамі: [n̪ ɦ ɑ mʲ i] * діяв: [dʲ i j ɑ u] |
Occurrences: 13,975 Examples: * гусар: [ɦ u s̪ ɐ ɾ] * діяв: [dʲ i j ɑ u] * зруб: [z̪ ɾ u b] * давні: [d̪ ɑ u ɲ i] |
|||
Occurrences: 24,405 Examples: * налию: [n̪ ɐ l ɪ j ʊ] * схилі: [s̪ x ɪ ʎ i] * нести: [n̪ e s̪ t̪ ɪ] * мойри: [m ɔ j ɾ ɪ] |
Occurrences: 17,156 Examples: * буває: [b ʊ ʋ ɑ j e] * учора: [ʊ tʃ ɔ ɾ ɐ] * ззаду: [z̪ː ɑ d̪ ʊ] * налию: [n̪ ɐ l ɪ j ʊ] |
||||
Close-Mid |
Occurrences: 30,731 Examples: * нічне: [ɲ i tʃ n̪ e] * буває: [b ʊ ʋ ɑ j e] * нести: [n̪ e s̪ t̪ ɪ] * ймемо: [i m e m ɔ] |
Occurrences: 2,551 Examples: * дворі: [d̪ ʋ o ɾʲ i] * водій: [ʋ o dʲ i i] * окріп: [o k ɾʲ i p] * родів: [ɾ o dʲ i u] |
|||
Open-Mid |
Occurrences: 7,519 Examples: * п'єте: [p j ɛ t̪ e] * нєма: [n̪ ɛ m ɐ] * зніме: [zʲ ɲ i m ɛ] * глек: [ɦ l ɛ k] |
Occurrences: 40,284 Examples: * офіс: [ɔ fʲ i s̪] * учора: [ʊ tʃ ɔ ɾ ɐ] * томас: [t̪ ɔ m ɑ s̪] * зночі: [z̪ n̪ ɔ tʃʲ i] |
|||
Occurrences: 33,085 Examples: * гусар: [ɦ u s̪ ɐ ɾ] * захар: [z̪ ɐ x ɑ ɾ] * учора: [ʊ tʃ ɔ ɾ ɐ] * налию: [n̪ ɐ l ɪ j ʊ] |
|||||
Open |
Occurrences: 22,846 Examples: * буває: [b ʊ ʋ ɑ j e] * нгамі: [n̪ ɦ ɑ mʲ i] * захар: [z̪ ɐ x ɑ ɾ] * томас: [t̪ ɔ m ɑ s̪] |