Vietnamese (Ho Chi Minh City) MFA dictionary v3.0.0#
@techreport{mfa_vietnamese_ho_chi_minh_city_mfa_dictionary_2024,
author={McAuliffe, Michael and Sonderegger, Morgan},
title={Vietnamese (Ho Chi Minh City) MFA dictionary v3.0.0},
address={\url{https://mfa-models.readthedocs.io/pronunciation dictionary/Vietnamese/Vietnamese (Ho Chi Minh City) MFA dictionary v3_0_0.html}},
year={2024},
month={Mar},
}
|
Installation#
Install from the MFA command line:
mfa model download dictionary vietnamese_ho_chi_minh_city_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/vietnamese/mfa/Vietnamese (Ho Chi Minh City) MFA dictionary v3_0_0.dict).
Intended use#
This dictionary is intended for forced alignment of Vietnamese transcripts.
This dictionary uses the MFA phone set for Vietnamese, and was used in training the Vietnamese 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 |
Alveolar |
Retroflex |
Palatal |
Velar |
Glottal |
---|---|---|---|---|---|---|---|
Nasal |
Occurrences: 511 Examples: * khẳm: [x a˨˩˦ m] * nám: [n aː˦˥ m] * mua: [m uə˨˨] * cảm: [k aː˨˩˦ m] |
Occurrences: 339 Examples: * bền: [ɓ əː˨˩ n] * gìn: [j ɨ˨˩ n] * nàng: [n aː˨˩ ŋ] * bênh: [ɓ əː˨˨ n] |
Occurrences: 150 Examples: * nhản: [ɲ aː˨˩˦ ŋ] * nhối: [ɲ oː˦˥ j] * nhàng: [ɲ aː˨˩ ŋ] * nhập: [ɲ ə˨˩˨ p̚] |
Occurrences: 935 Examples: * rợn: [r əː˨˩˨ ŋ] * ngắm: [ŋ a˦˥ m] * ngông: [ŋ ɔ˨˨ ŋm] * khiển: [x iː˨˩˦ ŋ] |
|||
Stop Plain |
Occurrences: 4 Examples: * pin: [p ɨ˨˨ n] |
Occurrences: 248 Examples: * từng: [t ɨ˨˩ ŋ] * tính: [t ɨ˦˥ n] * tắn: [t a˦˥ ŋ] * toa: [t w aː˨˨] |
Occurrences: 139 Examples: * triết: [ʈ iː˦˥] * trăng: [ʈ a˨˨ ŋ] * triền: [ʈ iː˨˩ ŋ] * tru: [ʈ u˨˨] |
Occurrences: 198 Examples: * chiên: [c iː˨˨ ŋ] * chống: [c ɔ˦˥ ŋm] * chước: [c ɨː˦˥ k̚] * chọt: [c ɔː˨˩˨ k̚] |
Occurrences: 266 Examples: * cày: [k a˨˩ j] * cầu: [k ə˨˩ w] * của: [k uə˨˩˦] * cấu: [k ə˦˥ w] Occurrences: 98 Examples: * gặp: [ɡ a˨˩˨ p̚] * gót: [ɡ ɔː˦˥ k̚] * gói: [ɡ ɔː˦˥ j] * gây: [ɡ ə˨˨ j] |
Occurrences: 146 Examples: * ứng: [ʔ ɨ˦˥ ŋ] * ước: [ʔ ɨː˦˥ k̚] * óng: [ʔ ɔ˦˥ ŋm] * inh: [ʔ ɨ˨˨ n] |
|
Aspirated |
Occurrences: 212 Examples: * thậm: [tʰ ə˨˩˨ m] * thưa: [tʰ ɨə˨˨] * tháng: [tʰ aː˦˥ ŋ] * thận: [tʰ ə˨˩˨ ŋ] |
||||||
Implosive |
Occurrences: 225 Examples: * bắt: [ɓ a˦˥] * bủa: [ɓ uə˨˩˦] * bào: [ɓ aː˨˩ w] * bầm: [ɓ ə˨˩ m] |
Occurrences: 223 Examples: * đắc: [ɗ a˦˥ k̚] * điển: [ɗ iː˨˩˦ ŋ] * đặc: [ɗ a˨˩˨ k̚] * đuối: [ɗ uː˦˥ j] |
|||||
Unreleased |
Occurrences: 137 Examples: * ngợp: [ŋ əː˨˩˨ p̚] * dép: [j ɛː˦˥ p̚] * búp: [ɓ ɨ˦˥ p̚] * ngáp: [ŋ aː˦˥ p̚] |
Occurrences: 102 Examples: * quýt: [w ɨ˦˥ t̚] * trách: [ʈ a˦˥ t̚] * ếch: [ʔ əː˦˥ t̚] * khít: [x ɨ˦˥ t̚] |
Occurrences: 373 Examples: * đợt: [ɗ əː˨˩˨ k̚] * nhắc: [ɲ a˦˥ k̚] * bát: [ɓ aː˦˥ k̚] * vẹt: [v ɛː˨˩˨ k̚] |
||||
Sibilant |
Occurrences: 281 Examples: * xấu: [s ə˦˥ w] * sức: [s ɨ˦˥ k̚] * sao: [s aː˨˨ w] * xương: [s ɨː˨˨ ŋ] Occurrences: 1 Examples: |
||||||
Fricative |
Occurrences: 99 Examples: * phấn: [f ə˦˥ ŋ] * phét: [f ɛː˦˥ k̚] * phùng: [f u˨˩ ŋm] * phó: [f ɔː˦˥] Occurrences: 165 Examples: * văng: [v a˨˨ ŋ] * vết: [v əː˦˥] * vẹn: [v ɛː˨˩˨ ŋ] * vươn: [v ɨː˨˨ ŋ] |
Occurrences: 229 Examples: * hân: [h ə˨˨ ŋ] * hợp: [h əː˨˩˨] * hoan: [h w aː˨˨ ŋ] * hán: [h aː˦˥ ŋ] |
|||||
Approximant |
Occurrences: 610 Examples: * xạo: [s aː˨˩˨ w] * chóe: [c w ɛː˦˥] * quen: [w ɛː˨˨ ŋ] * hoạn: [h w aː˨˩˨ ŋ] |
Occurrences: 692 Examples: * cấy: [k ə˦˥ j] * cài: [k aː˨˩ j] * sôi: [s oː˨˨ j] * hai: [h aː˨˨ j] |
|||||
Trill |
Occurrences: 150 Examples: * rút: [r u˦˥ kp] * rác: [r aː˦˥ k̚] * răn: [r a˨˨ ŋ] * rửa: [r ɨə˨˩˦] |
||||||
Lateral |
Occurrences: 259 Examples: * lầy: [l ə˨˩ j] * lùa: [l uə˨˩] * lươn: [l ɨː˨˨ ŋ] * linh: [l ɨ˨˨ n] |
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: 166 Examples: * quí: [w i˦˥] * quĩ: [w i˨˩˦] * tri: [ʈ i˨˨] * thi: [tʰ i˨˨] Occurrences: 270 Examples: * biện: [ɓ iː˨˩˨ ŋ] * uyển: [w iː˨˩˦ ŋ] * chiếc: [c iː˦˥] * quyết: [w iː˦˥] |
Occurrences: 336 Examples: * cựu: [k ɨ˨˩˨ w] * đít: [ɗ ɨ˦˥ t̚] * quýnh: [w ɨ˦˥ n] * ngưng: [ŋ ɨ˨˨ ŋ] Occurrences: 135 Examples: * lược: [l ɨː˨˩˨ k̚] * dương: [j ɨː˨˨ ŋ] * nướng: [n ɨː˦˥ ŋ] * lưới: [l ɨː˦˥ j] |
Occurrences: 264 Examples: * thủy: [tʰ u˨˩˦ j] * tụi: [t u˨˩˨ j] * thun: [tʰ u˨˨ ŋm] * tùy: [t u˨˩ j] Occurrences: 81 Examples: * đuôi: [ɗ uː˨˨ j] * muộn: [m uː˨˩˨ ŋ] * cuốn: [k uː˦˥ ŋ] * ruồi: [r uː˨˩ j] |
||
Close-Mid |
Occurrences: 0 Examples: Occurrences: 120 Examples: * chê: [c eː˨˨] * đềm: [ɗ eː˨˩ m] * tệp: [t eː˨˩˨ p̚] * giễu: [j eː˨˩˦ w] |
Occurrences: 72 Examples: * hổn: [h o˨˩˦ ŋ] * côn: [k o˨˨ ŋ] * hỗn: [h o˨˩˦ ŋ] * đồn: [ɗ o˨˩ ŋ] Occurrences: 165 Examples: * ngô: [ŋ oː˨˨] * tội: [t oː˨˩˨ j] * trồ: [ʈ oː˨˩] * phổ: [f oː˨˩˦] |
|||
Occurrences: 390 Examples: * vật: [v ə˨˩˨ k̚] * tấu: [t ə˦˥ w] * tấc: [t ə˦˥ k̚] * lần: [l ə˨˩ ŋ] Occurrences: 263 Examples: * ngơi: [ŋ əː˨˨ j] * mời: [m əː˨˩ j] * dời: [j əː˨˩ j] * nền: [n əː˨˩ n] |
|||||
Open-Mid |
Occurrences: 0 Examples: Occurrences: 243 Examples: * nhép: [ɲ ɛː˦˥ p̚] * ken: [k ɛː˨˨ ŋ] * đẹp: [ɗ ɛː˨˩˨ p̚] * bẽn: [ɓ ɛː˨˩˦ ŋ] |
Occurrences: 180 Examples: * cổng: [k ɔ˨˩˦ ŋm] * xong: [s ɔ˨˨ ŋm] * chồng: [c ɔ˨˩ ŋm] * vóc: [v ɔ˦˥ kp] Occurrences: 203 Examples: * trỏ: [ʈ ɔː˨˩˦] * bọt: [ɓ ɔː˨˩˨] * thon: [tʰ ɔː˨˨ ŋ] * trọt: [ʈ ɔː˨˩˨ k̚] |
|||
Open |
Occurrences: 429 Examples: * lạch: [l a˨˩˨] * ngành: [ŋ a˨˩ n] * vắt: [v a˦˥ k̚] * khắc: [x a˦˥] Occurrences: 727 Examples: * láo: [l aː˦˥ w] * vài: [v aː˨˩ j] * trà: [ʈ aː˨˩] * gan: [ɡ aː˨˨ ŋ] |
Diphthongs#
iə
uə
ɨə
Tones#
˦˥
˨˨
˨˩
˨˩˦
˨˩˨