Vietnamese (Hanoi) MFA dictionary v2.0.0#

Maintainer: Montreal Forced Aligner
Language: Vietnamese
Dialect: N/A
Phone set: MFA
Number of words: 4,810
Phones: aː˦˥ aː˨ˀ˥ aː˨˦ aː˨˨ aː˨˩ aː˨˩ˀ aː˨˩˨ a˦˥ a˨ˀ˥ a˨˦ a˨˨ a˨˩ a˨˩ˀ a˨˩˨ c eː˦˥ eː˨ˀ˥ eː˨˦ eː˨˨ eː˨˩ eː˨˩ˀ eː˨˩˨ e˦˥ e˨ˀ˥ e˨˨ e˨˩ e˨˩ˀ e˨˩˨ f h iə˦˥ iə˨ˀ˥ iə˨˦ iə˨˨ iə˨˩ iə˨˩ˀ iə˨˩˨ iː˦˥ iː˨ˀ˥ iː˨˦ iː˨˨ iː˨˩ iː˨˩ˀ iː˨˩˨ i˦˥ i˨ˀ˥ i˨˦ i˨˨ i˨˩ i˨˩ˀ i˨˩˨ j k kp l m n oː˦˥ oː˨ˀ˥ oː˨˦ oː˨˨ oː˨˩ oː˨˩ˀ oː˨˩˨ o˦˥ o˨ˀ˥ o˨˦ o˨˨ o˨˩ o˨˩ˀ o˨˩˨ p s t tɕ tʰ uə˦˥ uə˨ˀ˥ uə˨˦ uə˨˨ uə˨˩ uə˨˩ˀ uə˨˩˨ uː˦˥ uː˨ˀ˥ uː˨˦ uː˨˨ uː˨˩ uː˨˩ˀ uː˨˩˨ u˦˥ u˨ˀ˥ u˨˦ u˨˨ u˨˩ u˨˩ˀ u˨˩˨ v w x z ŋ ŋm ɓ ɔː˦˥ ɔː˨ˀ˥ ɔː˨˦ ɔː˨˨ ɔː˨˩ ɔː˨˩ˀ ɔː˨˩˨ ɔ˦˥ ɔ˨ˀ˥ ɔ˨˦ ɔ˨˨ ɔ˨˩ ɔ˨˩ˀ ɔ˨˩˨ ɗ əː˦˥ əː˨ˀ˥ əː˨˦ əː˨˨ əː˨˩ əː˨˩ˀ əː˨˩˨ ə˦˥ ə˨ˀ˥ ə˨˦ ə˨˨ ə˨˩ ə˨˩ˀ ə˨˩˨ ɛː˦˥ ɛː˨ˀ˥ ɛː˨˦ ɛː˨˨ ɛː˨˩ ɛː˨˩ˀ ɛː˨˩˨ ɣ ɨə˦˥ ɨə˨ˀ˥ ɨə˨˦ ɨə˨˨ ɨə˨˩ ɨə˨˩ˀ ɨə˨˩˨ ɨː˨ˀ˥ ɨː˨˦ ɨː˨˨ ɨː˨˩ˀ ɨː˨˩˨ ɨ˦˥ ɨ˨ˀ˥ ɨ˨˦ ɨ˨˨ ɨ˨˩ ɨ˨˩ˀ ɨ˨˩˨ ɲ ʔ
License: CC BY 4.0
Compatible MFA version: v2.0.0
Citation:

@techreport{mfa_vietnamese_hanoi_mfa_dictionary_2022,
	author={McAuliffe, Michael and Sonderegger, Morgan},
	title={Vietnamese (Hanoi) MFA dictionary v2.0.0},
	address={\url{https://mfa-models.readthedocs.io/pronunciation dictionary/Vietnamese/Vietnamese (Hanoi) MFA dictionary v2_0_0.html}},
	year={2022},
	month={Mar},
}

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.

G2P models

Vietnamese (Hanoi) MFA G2P model v2.0.0

Installation#

Install from the MFA command line:

mfa model download dictionary vietnamese_hanoi_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 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	Palatal	Velar	Glottal
Nasal	m Occurrences: 654 Examples: * nghìm: [ŋ iː˨˨ m] * mập: [m ə˨˩ p] * mứt: [m ɨ˦˥ t] * bờm: [ɓ əː˨˨ m]		n Occurrences: 860 Examples: * tròn: [tɕ ɔː˨˨ n] * tuyển: [t w iə˨˩˨ n] * đoạn: [ɗ w aː˨˩ˀ n] * cản: [k aː˨˩˨ n]	ɲ Occurrences: 354 Examples: * vểnh: [v e˨˩˨ ɲ] * lạnh: [l a˨˩ˀ ɲ] * mánh: [m a˨˦ ɲ] * nhung: [ɲ u˨˨ ŋm]	ŋ Occurrences: 567 Examples: * tràng: [tɕ aː˨˨ ŋ] * nghé: [ŋ ɛː˨˦] * nghìm: [ŋ iː˨˨ m] * chẳng: [tɕ a˨˩˨ ŋ]
Stop Plain	p Occurrences: 197 Examples: * mập: [m ə˨˩ p] * hẹp: [h ɛː˨˩ p] * giúp: [z uː˦˥ p] * thép: [tʰ ɛː˦˥ p]		t Occurrences: 674 Examples: * quát: [k w aː˦˥ t] * tuyển: [t w iə˨˩˨ n] * khoét: [x w ɛː˦˥ t] * vuốt: [v uə˦˥ t]	c Occurrences: 82 Examples: * ách: [ʔ a˦˥ c] * vích: [v i˦˥ c] * oách: [w a˦˥ c] * trách: [tɕ a˦˥ c]	k Occurrences: 543 Examples: * quát: [k w aː˦˥ t] * cãi: [k aː˨ˀ˥ j] * cản: [k aː˨˩˨ n] * kiến: [k iə˨˦ n]	ʔ Occurrences: 189 Examples: * ổng: [ʔ o˨˩˨ ŋm] * yên: [ʔ iə˨˨ n] * uôn: [ʔ uə˨˨ n] * ách: [ʔ a˦˥ c]
Implosive	ɓ Occurrences: 266 Examples: * bêu: [ɓ eː˨˨ w] * bờm: [ɓ əː˨˨ m] * biểu: [ɓ iə˨˩˨ w] * biển: [ɓ iə˨˩˨ n]		ɗ Occurrences: 252 Examples: * đoạn: [ɗ w aː˨˩ˀ n] * đường: [ɗ ɨə˨˨ ŋ] * đăm: [ɗ a˨˨ m] * đắn: [ɗ a˨˦ n]
Affricate				tɕ Occurrences: 392 Examples: * chày: [tɕ a˨˨ j] * tràng: [tɕ aː˨˨ ŋ] * tròn: [tɕ ɔː˨˨ n] * chẳng: [tɕ a˨˩˨ ŋ]
Sibilant			s Occurrences: 337 Examples: * xoài: [s w aː˨˨ j] * sởi: [s əː˨˩˨ j] * sọt: [s ɔː˨˩ t] * sảo: [s aː˨˩˨ w] z Occurrences: 501 Examples: * dẫm: [z ə˨ˀ˥ m] * gioan: [z w aː˨˨ n] * ruộng: [z uə˨˩ˀ ŋ] * giởi: [z əː˨˩˨ j]
Fricative		f Occurrences: 121 Examples: * phanh: [f a˨˨ ɲ] * phình: [f i˨˨ ɲ] * phây: [f ə˨˨ j] * phè: [f ɛː˨˨] v Occurrences: 188 Examples: * vuốt: [v uə˦˥ t] * vểnh: [v e˨˩˨ ɲ] * vịn: [v iː˨˩ˀ n] * vích: [v i˦˥ c]				h Occurrences: 256 Examples: * hao: [h aː˨˨ w] * hẹp: [h ɛː˨˩ p] * học: [h ɔ˨˩ kp] * hồi: [h oː˨˨ j]
Approximant	w Occurrences: 779 Examples: * quát: [k w aː˦˥ t] * hao: [h aː˨˨ w] * tuyển: [t w iə˨˩˨ n] * khoét: [x w ɛː˦˥ t]			j Occurrences: 602 Examples: * chày: [tɕ a˨˨ j] * xoài: [s w aː˨˨ j] * sởi: [s əː˨˩˨ j] * cãi: [k aː˨ˀ˥ j]
Lateral			l Occurrences: 290 Examples: * lạnh: [l a˨˩ˀ ɲ] * lúa: [l uə˨˦] * lẫy: [l ə˨ˀ˥ j] * lưng: [l ɨ˨˨ ŋ]

Vowels#

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

	Front	Central	Back
Close	i Occurrences: 98 Examples: * thịnh: [tʰ i˨˩ˀ ɲ] * chỉnh: [tɕ i˨˩˨ ɲ] * định: [ɗ i˨˩ˀ ɲ] * chích: [tɕ i˦˥ c] iː Occurrences: 282 Examples: * rít: [z iː˦˥ t] * nhịp: [ɲ iː˨˩ p] * thị: [tʰ iː˨˩ˀ] * thuỵ: [tʰ w iː˨˩ˀ]	ɨ Occurrences: 119 Examples: * ửng: [ʔ ɨ˨˩˨ ŋ] * sực: [s ɨ˨˩ k] * dức: [z ɨ˦˥ k] * đứng: [ɗ ɨ˨˦ ŋ] ɨː Occurrences: 55 Examples: * khứ: [x ɨː˨˦] * nhử: [ɲ ɨː˨˩˨] * khử: [x ɨː˨˩˨] * chữ: [tɕ ɨː˨ˀ˥]	u Occurrences: 109 Examples: * nhung: [ɲ u˨˨ ŋm] * dụng: [z u˨˩ˀ ŋm] * rúc: [z u˦˥ kp] * trũng: [tɕ u˨ˀ˥ ŋm] uː Occurrences: 267 Examples: * rụt: [z uː˨˩ t] * hụi: [h uː˨˩ˀ j] * hụt: [h uː˨˩ t] * giúp: [z uː˦˥ p]

Close-Mid	e Occurrences: 38 Examples: * tênh: [t e˨˨ ɲ] * mệnh: [m e˨˩ˀ ɲ] * chênh: [tɕ e˨˨ ɲ] * ễnh: [ʔ e˨ˀ˥ ɲ] eː Occurrences: 180 Examples: * vệt: [v eː˨˩ t] * đệm: [ɗ eː˨˩ˀ m] * khuê: [x w eː˨˨] * nhễ: [ɲ eː˨ˀ˥]		o Occurrences: 103 Examples: * bổng: [ɓ o˨˩˨ ŋm] * thốc: [tʰ o˦˥ kp] * sổng: [s o˨˩˨ ŋm] * độc: [ɗ o˨˩ kp] oː Occurrences: 268 Examples: * chột: [tɕ oː˨˩ t] * ngôn: [ŋ oː˨˨ n] * đỗi: [ɗ oː˨ˀ˥ j] * đốm: [ɗ oː˨˦ m]
		ə Occurrences: 428 Examples: * dần: [z ə˨˨ n] * đần: [ɗ ə˨˨ n] * tuấn: [t w ə˨˦ n] * tẩm: [t ə˨˩˨ m] əː Occurrences: 246 Examples: * chơi: [tɕ əː˨˨ j] * đớp: [ɗ əː˦˥ p] * nhở: [ɲ əː˨˩˨] * hơn: [h əː˨˨ n]
Open-Mid	ɛː Occurrences: 307 Examples: * bèn: [ɓ ɛː˨˨ n] * ghẹo: [ɣ ɛː˨˩ˀ w] * ghém: [ɣ ɛː˨˦ m] * nghẽn: [ŋ ɛː˨ˀ˥ n]		ɔ Occurrences: 108 Examples: * vong: [v ɔ˨˨ ŋm] * dóc: [z ɔ˦˥ kp] * võng: [v ɔ˨ˀ˥ ŋm] * học: [h ɔ˨˩ kp] ɔː Occurrences: 271 Examples: * đón: [ɗ ɔː˨˦ n] * gọi: [ɣ ɔː˨˩ˀ j] * móp: [m ɔː˦˥ p] * chói: [tɕ ɔː˨˦ j]

Open		a Occurrences: 481 Examples: * gặm: [ɣ a˨˩ˀ m] * nẳm: [n a˨˩˨ m] * ách: [ʔ a˦˥ c] * trách: [tɕ a˦˥ c] aː Occurrences: 783 Examples: * sái: [s aː˨˦ j] * sảo: [s aː˨˩˨ w] * xạc: [s aː˨˩ k] * nhát: [ɲ aː˦˥ t]

Diphthongs#

iə
uə
ɨə

Tones#

˦˥
˨ˀ˥
˨˦
˨˨
˨˩
˨˩ˀ
˨˩˨