Ukrainian MFA dictionary v3.0.0#
@techreport{mfa_ukrainian_mfa_dictionary_2024,
author={McAuliffe, Michael and Sonderegger, Morgan},
title={Ukrainian MFA dictionary v3.0.0},
address={\url{https://mfa-models.readthedocs.io/pronunciation dictionary/Ukrainian/Ukrainian MFA dictionary v3_0_0.html}},
year={2024},
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](https://raw.githubusercontent.com/MontrealCorpusTools/mfa-models/main/dictionary/ukrainian/mfa/Ukrainian MFA dictionary v3_0_0.dict).
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: 11,729 Examples: * ману: [m ɐ n̪ ʊ] * матки: [m ɑ t̪ k ɪ] * мирні: [m ɪ ɾ ɲ i] * многі: [m n̪ ɔ ʝ i] Occurrences: 1,313 Examples: * мізку: [mʲ i z̪ k ʊ] * місіс: [mʲ i sʲ i s̪] * міс: [mʲ i s̪] * німі: [ɲ i mʲ i] Occurrences: 8 Examples: |
Occurrences: 17,447 Examples: * вікон: [ʋʲ i k ɔ n̪] * дивну: [d̪ ɪ ʋ n̪ ʊ] * бощан: [b ɔ ʃ tʃ ɑ n̪] * інший: [i n̪ ʃ e i] Occurrences: 378 Examples: * панно: [p ɑ n̪ː ɔ] * денне: [d̪ e n̪ː e] * сонно: [s̪ ɔ ɔ n̪ː ɔ] * ганно: [ɦ ɑ n̪ː ɔ] |
Occurrences: 5,049 Examples: * кузню: [k u zʲ ɲ ʊ] * нюх: [ɲ u x] * ніжці: [ɲ i ʒ tsʲ i] * нього: [ɲ ɔ ɦ ɔ] Occurrences: 1,021 Examples: * тінню: [tʲ i ɲː ʊ] * кінні: [c i ɲː i] * хінні: [ç i ɲː i] * рання: [ɾ ɑ ɲː ɐ] |
|||||
Stop |
Occurrences: 13,843 Examples: * поти: [p ɔ t̪ ɪ] * випив: [ʋ ɪ p ɪ ʋ] * плай: [p l ɑ i] * шапка: [ʃ ɑ p k ɐ] Occurrences: 1,388 Examples: * спір: [sʲ pʲ i ɾ] * співу: [sʲ pʲ i ʋ ʊ] * пірка: [pʲ i ɾ k ɐ] * спід: [sʲ pʲ i d̪] Occurrences: 1 Examples: Occurrences: 6,027 Examples: * блиск: [b l ɪ s̪ k] * буцім: [b ʊ tsʲ i m] * абрам: [ɐ b ɾ ɐ m] * баня: [b ɑ ɲ ɐ] Occurrences: 774 Examples: * дубі: [d̪ ʊ bʲ i] * обіді: [ɔ bʲ i dʲ i] * небіж: [n̪ ɛ bʲ i ʒ] * бік: [bʲ i k] Occurrences: 3 Examples: Occurrences: 1 Examples: |
Occurrences: 15,354 Examples: * тісто: [tʲ i s̪ t̪ ɔ] * гатку: [ɦ ɑ t̪ k ʊ] * хат: [x ɑ t̪] * ост: [ɔ s̪ t̪] Occurrences: 16 Examples: Occurrences: 9,899 Examples: * вход: [ʋ x ɔ d̪] * ридав: [ɾ ɪ d̪ ɑ ʋ] * ядру: [j ɐ d̪ ɾ ʊ] * вадим: [ʋ ɑ d̪ ɪ m] Occurrences: 105 Examples: * оддам: [ɔ d̪ː ɐ m] * оддає: [ɔ d̪ː ɑ j e] * оддай: [ɔ d̪ː ɑ i] * оддав: [o d̪ː ɑ ʋ] |
Occurrences: 0 Examples: Occurrences: 5,220 Examples: * ідуть: [i d̪ u tʲ] * втіха: [ʋ tʲ i x ɐ] * дасть: [d̪ ɐ sʲ tʲ] * ждуть: [ʒ d̪ u tʲ] Occurrences: 138 Examples: * пуття: [p ʊ tʲː ɑ] * виття: [ʋ e tʲː ɑ] * життє: [ʒ e tʲː ɛ] * шиття: [ʃ e tʲː ɑ] Occurrences: 0 Examples: Occurrences: 2,199 Examples: * діво: [dʲ i ʋ ɔ] * радіо: [ɾ ɑ dʲ i ɔ] * буді: [b u dʲ i] * дню: [dʲ ɲ ʊ] Occurrences: 45 Examples: * міддю: [mʲ i dʲː ʊ] * суддю: [s̪ ʊ dʲː u] * суддя: [s̪ ʊ dʲː ɑ] * судді: [s̪ ʊ dʲː i] |
Occurrences: 970 Examples: * якім: [j ɑ c i m] * кінні: [c i ɲː i] * кіш: [c i ʃ] * меткі: [m ɛ t̪ ɔ c i] Occurrences: 1 Examples: Occurrences: 3 Examples: * ґміни: [ɟ mʲ i n̪ ɪ] * ґніт: [ɟ ɲ i t̪] |
Occurrences: 14,664 Examples: * кимсь: [k ɪ m sʲ] * литку: [l e t̪ k ʊ] * дудка: [d̪ u d̪ k ɐ] * кохаю: [k ɔ x ɑ j ʊ] Occurrences: 7 Examples: * мекка: [m ɛ kː ɐ] * мекку: [m ɛ kː ʊ] Occurrences: 81 Examples: * манґи: [m ɐ n̪ ɡ ɪ] * ґрунт: [ɡ ɾ u n̪ t̪] * ґвалт: [ɡ ʋ ɑ l t̪] * дзиґа: [d̪z̪ ɪ ɡ ɐ] |
|||
Affricate |
Occurrences: 723 Examples: * цапів: [t̪s̪ ɑ pʲ i ʋ] * отцем: [ɔ d̪ t̪s̪ e m] * оцей: [ɔ t̪s̪ ɛ i] * цапи: [t̪s̪ ɑ p ɪ] Occurrences: 2 Examples: * цска: [t̪s̪ː k ɐ] Occurrences: 346 Examples: * відси: [ʋʲ i d̪z̪ s̪ ɪ] * ґудзь: [ɡ u d̪z̪] * дззи: [d̪z̪ z̪ ɪ] * дзиґа: [d̪z̪ ɪ ɡ ɐ] Occurrences: 2 Examples: * дзз: [d̪z̪ː] |
Occurrences: 0 Examples: Occurrences: 2,158 Examples: * борці: [b ɔ ɾ tsʲ i] * місці: [mʲ i sʲ tsʲ i] * плець: [p l e tsʲ] * ціпок: [tsʲ i p ɔ k] Occurrences: 143 Examples: * гудця: [ɦ ʊ tsʲː ɐ] * сітці: [sʲ i ɔ tsʲː i] * кітці: [c i ɔ tsʲː i] * річці: [ɾʲ i tsʲː i] Occurrences: 0 Examples: Occurrences: 71 Examples: * дзьоб: [dzʲ ɔ b] * дзвін: [dzʲ ʋʲ i n̪] |
Occurrences: 6,504 Examples: * кучми: [k u tʃ m ɪ] * січ: [sʲ i tʃ] * дещо: [d̪ e ʃ tʃ ɔ] * чуб: [tʃ ʊ b] Occurrences: 395 Examples: * чітку: [tʃʲ i t̪ k ʊ] * онучі: [o n̪ u tʃʲ i] * мощі: [m ɔ ʃ tʃʲ i] * утечі: [ʊ t̪ e tʃʲ i] Occurrences: 24 Examples: * рітчі: [ɾʲ i tʃʲː i] * віччю: [ʋʲ i tʃʲː ʊ] * ніччю: [ɲ i tʃʲː ʊ] * матчі: [m ɐ tʃʲː i] Occurrences: 71 Examples: * матчу: [m ɐ tʃː ʊ] * лучче: [l u tʃː e] * одчай: [ɔ tʃː ɐ i] * матч: [m ɐ tʃː] Occurrences: 444 Examples: * джуді: [dʒ ʊ dʲ i] * імідж: [i mʲ i dʒ] * бджіл: [b dʒ i l] * джерю: [dʒ e ɾʲ ʊ] |
|||||
Sibilant |
Occurrences: 11,279 Examples: * савки: [s̪ ɐ ʋ k ɪ] * цар: [t̪s̪ ɑ ɾ] * сало: [s̪ ɐ l ɔ] * осені: [ɔ s̪ ɛ ɲ i] Occurrences: 19 Examples: * ссати: [s̪ː ɑ t̪ ɪ] * ссала: [s̪ː ɐ l ɐ] * руссю: [ɾ u s̪ː ʊ] * улісс: [ʊ ʎ i s̪ː] Occurrences: 8,333 Examples: * низом: [n̪ ɪ z̪ ɔ m] * заким: [z̪ ɐ k ɪ m] * збити: [z̪ b ɪ t̪ ɪ] * зона: [z̪ ɔ n̪ ɐ] Occurrences: 15 Examples: * дзз: [d̪z̪ː] * ззаду: [z̪ː ɑ d̪ ʊ] * ззаді: [z̪ː ɑ dʲ i] |
Occurrences: 0 Examples: Occurrences: 9,211 Examples: * сіра: [sʲ i ɾ ɐ] * якось: [j ɐ k ɔ sʲ] * уся: [u sʲ ɐ] * сіяв: [sʲ i j ɑ ʋ] Occurrences: 104 Examples: * нісся: [ɲ i sʲː ɐ] * россю: [ɾ o sʲː u] * отся: [ɔ sʲː ɐ] * отсі: [ɔ sʲː i] Occurrences: 0 Examples: Occurrences: 1,388 Examples: * зятю: [zʲ ɑ tʲː ʊ] * князі: [k ɲ ɑ zʲ i] * зняті: [zʲ ɲ ɑ tʲ i] * разів: [ɾ ɑ zʲ i ʋ] Occurrences: 8 Examples: |
Occurrences: 5,024 Examples: * груша: [ɦ ɾ u ʃ ɐ] * шити: [ʃ ɪ t̪ ɪ] * душив: [d̪ u ʃ ɪ ʋ] * шапку: [ʃ ɑ p k ʊ] Occurrences: 134 Examples: * довші: [d̪ ɔ ʋ ʃʲ i] * нашій: [n̪ ɑ ʃʲ iː] * нашім: [n̪ ɑ ʃʲ i m] * миші: [m e ʃʲ i] Occurrences: 2,866 Examples: * враже: [ʋ ɾ ɑ ʒ e] * жадаю: [ʒ ɑ d̪ ɐ j ʊ] * межа: [m e ʒ ɑ] * кожне: [k ɔ ʒ n̪ ɛ] Occurrences: 62 Examples: * жінку: [ʒʲ i n̪ k ʊ] * етажі: [e t̪ ɐ ʒʲ i] * жінці: [ʒʲ i ɲ tsʲ i] * жінка: [ʒʲ i n̪ k ɐ] Occurrences: 17 Examples: |
|||||
Fricative |
Occurrences: 568 Examples: * шофер: [ʃ ɔ f ɛ ɾ] * фронт: [f ɾ ɔ n̪ t̪] * флоті: [f l ɔ tʲ i] * фойє: [f ɔ i e] Occurrences: 214 Examples: * фіно: [fʲ i n̪ ɔ] * фін: [fʲ i n̪] * фільм: [fʲ i ʎ m] * шафі: [ʃ ɑ fʲ i] Occurrences: 2 Examples: * моффа: [m ɔ fː ɐ] |
Occurrences: 189 Examples: * ляхів: [ʎ ɑ ç i ʋ] * хід: [ç i d̪] * лихі: [l e ç i] * рахів: [ɾ ɐ ç i ʋ] Occurrences: 388 Examples: * гілля: [ʝ i ʎː ɐ] * убогі: [ʊ b ɔ ʝ i] * гілці: [ʝ i tsʲ i] * легіт: [l e ʝ i t̪] |
Occurrences: 7,031 Examples: * голка: [ɦ ɔ l k ɐ] * граба: [ɦ ɾ ɐ b ɐ] * смуга: [s̪ m u ɦ ɐ] * нього: [ɲ ɔ ɦ ɔ] |
|||||
Approximant |
Occurrences: 21,243 Examples: * сиваш: [s̪ e ʋ ɑ ʃ] * вид: [ʋ ɪ d̪] * кусав: [k ʊ s̪ ɑ ʋ] * весло: [ʋ ɛ s̪ l ɔ] Occurrences: 3,043 Examples: * лаві: [l ɑ ʋʲ i] * візу: [ʋʲ i z̪ ʊ] * навіс: [n̪ ɑ ʋʲ i s̪] * віри: [ʋʲ i ɾ ɪ] Occurrences: 17 Examples: * вві: [ʋʲː i] * ввів: [ʋʲː i ʋ] Occurrences: 46 Examples: * ввело: [ʋː e l ɔ] * вверх: [ʋː e ɾ x] * ввесь: [ʋː ɛ sʲ] * вволю: [ʋː ɔ ʎ ʊ] |
Occurrences: 8,529 Examples: * хомою: [x ɔ m ɔ j ʊ] * яєць: [j ɐ j e tsʲ] * ясів: [j ɑ sʲ i ʋ] * дії: [dʲ i j i] |
||||||
Tap |
Occurrences: 19,999 Examples: * ромен: [ɾ ɔ m ɛ n̪] * русі: [ɾ u sʲ i] * хрест: [x ɾ ɛ s̪ t̪] * орля: [ɔ ɾ l ɑ] Occurrences: 2,410 Examples: * пріск: [p ɾʲ i s̪ k] * зорі: [z̪ ɔ ɾʲ i] * рівну: [ɾʲ i ʋ n̪ ʊ] * тріо: [tʲ ɾʲ i ɔ] Occurrences: 6 Examples: * гаррі: [ɦ ɐ ɾʲː i] Occurrences: 4 Examples: * гурра: [ɦ ʊ ɾː ɐ] * ферро: [f ɛ ɾː ɔ] |
|||||||
Lateral |
Occurrences: 14,044 Examples: * скали: [s̪ k ɑ l ɪ] * столи: [s̪ t̪ ɔ l ɪ] * лихе: [l e x ɛ] * цілує: [tsʲ i l ʊ j e] Occurrences: 34 Examples: * халлу: [x ɐ lː ʊ] * волл: [ʋ ɔ lː] * аллен: [ɐ lː ɛ n̪] * халл: [x ɐ lː] |
Occurrences: 4,894 Examples: * лягла: [ʎ ɐ ɦ l ɐ] * шлюбу: [ʃ ʎ u b ʊ] * лім: [ʎ i m] * любку: [ʎ ʊ b k ʊ] Occurrences: 72 Examples: * валль: [ʋ ɐ ʎː] * сіллю: [sʲ i ʎː ʊ] * заллє: [z̪ ɐ ʎː ɛ] * гілля: [ʝ i ʎː ɐ] |
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: 24,363 Examples: * заїр: [z̪ ɐ j i ɾ] * іде: [i d̪ ɛ] * фіри: [fʲ i ɾ ɪ] * мішку: [mʲ i ʃ k ʊ] Occurrences: 1,084 Examples: * пійло: [pʲ iː l ɔ] * бійся: [bʲ iː sʲ ɐ] * копій: [k o pʲ iː] * лівій: [ʎ i ʋʲ iː] |
Occurrences: 5,020 Examples: * юного: [j u n̪ ɔ ɦ ɔ] * круп: [k ɾ u p] * людей: [ʎ u d̪ ɛ i] * юшка: [j u ʃ k ɐ] |
|||
Occurrences: 21,505 Examples: * рикав: [ɾ ɪ k ɐ ʋ] * замки: [z̪ ɑ m k ɪ] * мине: [m ɪ n̪ e] * миски: [m ɪ s̪ k ɪ] |
Occurrences: 14,771 Examples: * суне: [s̪ ʊ n̪ e] * новою: [n̪ ɔ ʋ ɔ j ʊ] * краму: [k ɾ ɐ m ʊ] * дурну: [d̪ ʊ ɾ n̪ ʊ] |
||||
Close-Mid |
Occurrences: 25,968 Examples: * серед: [s̪ ɛ ɾ e d̪] * миль: [m e ʎ] * себе: [s̪ e b ɛ] * каноє: [k ɐ n̪ ɔ j e] |
Occurrences: 2,120 Examples: * поміг: [p o mʲ i ɦ] * окуня: [o k u ɲ ɐ] * софія: [s̪ o fʲ i j ɐ] * щоці: [ʃ tʃ o tsʲ i] |
|||
Open-Mid |
Occurrences: 6,129 Examples: * своєю: [s̪ ʋ ɔ j ɛ j ʊ] * густе: [ɦ ʊ s̪ t̪ ɛ] * вєм: [ʋ j ɛ m] * щезне: [ʃ tʃ ɛ z̪ n̪ e] |
Occurrences: 34,471 Examples: * колу: [k ɔ l ʊ] * обоз: [ɔ b ɔ z̪] * тої: [t̪ ɔ j i] * діяло: [dʲ i j ɑ l ɔ] |
|||
Occurrences: 27,661 Examples: * гайда: [ɦ ɑ i d̪ ɐ] * марії: [m ɐ ɾʲ i j i] * воля: [ʋ ɔ ʎ ɐ] * лада: [l ɑ d̪ ɐ] |
|||||
Open |
Occurrences: 19,839 Examples: * марку: [m ɑ ɾ k ʊ] * вдача: [ʋ d̪ ɑ tʃ ɐ] * п'ядь: [p j ɑ dʲ] * хадід: [x ɑ dʲ i d̪] |